foohbarbaz
USB requires a lot more work and on Windows the API is atrocious. On Linux the API (libusb) quite a bit nicer, but still a bit of work.
The plug and play part, device naming and unique identifiers are a special "joy" of USB.
If serial port is sufficiently fast I'll take it any day over USB.
Somewhat related: I am amazed now much gossiping goes on in an entirely male work group behind other people's backs. Males are definitely worse than females in that regard. I would not be surpised if allegations were true. A bunch if primadonna devs can be nasty to work with. Just stay away and separate work and life.
http://militera.lib.ru/research/suvorov12/07.html
is spot on! Been there.
So, you are doing 84mph (speed limit being 75) and go from relatively bright lit area into a dark and a curve. A second of blindness inevitably follows.
Still, I would probably rather just have a higher divider between lanes in different directions, so that high beams can be just used most of the time.
Over the mountain passes in spring where the lane markings get worn down sometimes you just can't see anything w/o high beams (and risking to blind the oncoming traffic).
What you get in return is both stability and security, since you don't wait for vendor to release a patch and actually understand what the vulnerability is.
With wholesale patching, in fact, you can never be sure whether your system is secure with respect to all published vulnerabilities.
Another interesting detail is, that with services running built from sources often you end up with vulnerabilities not applicable to your configuration.
Oftentimes, you can just tweak the config instead of changing code (and potentially breaking running things).
Software updates is just a cop out for people that are too lazy to pay attention to security.
You only patch what you need to patch. Most of the time for every production service you end up building a custom version anyway. Patching does no good to those.
So, by patching you only bring potential harm and overhead of going through change control processes.
I do not accept the risk of waiting for some vendor to release a patch. If there's a hole, read the report, determine whether your config/build is vulnerable, rebuild.
Why would want to patch something you are not running or use?
If I am running a service facing the internet, it's custom built and patches would do it no good. Why would I wait for vendor to release a patch? If a service is external, I will watch out for vulnerabilities and rebuild ASAP before any patches are out. Besides, 90% of the time my custom build is not even vulnerable to a particular problem.
If I am NOT running a service, why would I care about patches for it?
Why would I wholesale patch a server anyway? If somebody breaks in and gets a local shell, all is lost anyway. If they are not in, they are dealing with externally facing services only, see above. There are specific and counted number of daemons on every machine.
This whole patch-update thing is misguided and for people that want assurances and no responsibility.
Navajos, who don't make eye contact or smile either also take some getting used to now.
It's cultural, but Russia is a harsh place, so it goes deeper.
The part about people in the street not being helpful to each other is also true.
Just watch those dash cam videos where people witness a horrible crash, drive around and keep moving. You can die in the street and nobody might care. It's cultural and it's not a good thing.
The mob mentality is also a lot stronger. Westerners are a lot more tolerant to people being different or choosing their own way. It is in fact encouraged to be different and individual. Russians on the other hand, have whole layers of culture dedicated to making sure everybody is not sticking out of the crowd and not being an individual.
The host names were invented so people won't have to memorize IP addresses. IT morons here with their "functional hostnames" have managed to defeat the purpose of hostnames, so that they are now harder to remember than IP addresses!
Please please please get a clue, use geography, mythology, Simpsons, anything but stupid "bisappjmpthatthislakdh000283".
Since it's outdoorsmen that use these maps, having no trails makes them worthless.
Here's what I do. Go to the USGS store and download older maps in (georeferenced) raster PDF format. These are hi res scans of real USGS quad maps. Just pick up anything older than 2012/2013. Use Acroread to print each map in poster mode 75% scale landscape to two 11x17" pages. Splice the pages and you have pretty much perfect USGS quad. I just came back from a hike in Wind Rivers and I had 14 USGS quads with me, covering all of the range.
Another option is this:
This site creates a seamless map out of the same USGS maps. You can print via PDF using correct scale and paper size. This is great for custom maps (although I still prefer quads as mentioned above).
To USGS: nice try, but until you add all layers present on real maps these are not any good.
It's a pretty primitive drug too. Such a simple and small molecule, might just as well sniff glue, it seems to me.
Tobacco is so much fun in comparison. Quit that cold turkey many years ago. Still like it though.
=== On NixOS, you do not need to be root to install software. In addition to the system-wide ‘profile’ (set of installed packages), all user have their own profile in which they can install packages. Nix allows multiple versions of a package to coexist, so different users can have different versions of the same package installed in their respective profiles. If two users install the same version of a package, only one copy will be built or downloaded, and Nix’s security model ensures that this is secure. Users cannot install setuid binaries. ===
Requirement to be administrator to install software is the root of all evils in managing OS.
As far as keeping everything in a special location (/nix or whatever), this reminds me SCO OpenServer 5: they had all the files somewhere in /var; /bin, /sbin and everything else were just symlinks. It did not work all that well.
I am actually quite opposed to the syntax sugar features like "getters/setters from C#", "events" and what not. Java the language is already large enough. IDEs do a decent job dealing with boilerplate code. Please, no new kludgy additions to the language.
Those people that need functional features could use Scala now. Jython is there for those not liking static typing.
Why don't we just leave Java alone?
True that in typical chaos that reigns in code written by a disjointed group of people it ends up being Exceptions and RuntimeExceptions all over the place anyway. But why should disciplined people suffer?
Please leave my checked exceptions alone. I like then very much. I have not dealt with C#, but did with C++ and I prefer the Java way.
The arguments against checked exceptions that I have seen go along the lines of leaking data types from unrelated code layers. That's probably valid, but I have not experienced it. The rest seems like just whining about being forced to write code to handle errors, instead of letting things just happen.
So, yes, I suspect strongly there was a good reason for Adria Richards to be offended.
The firing by her company and the reaction of the community just left me speechless. Just plain sad. Why the dev in question needed to be fired is also questionable to me.
Ashamed to belong to this crowd. Unfortunately.
My impression is that inappropriate jokes were specifically targeted at a woman seated nearby. In an all male company (of 13-year old youths) it just does not happen.
The people that did this deserved to be reminded that it was not acceptable (although not by firing).
Now, firing Adria Richards is entirely beyond my understanding. Seriously, WTF? Well, this whole story is sad. Blame the messenger now?
If the OS is intact, the job of checking whether a user environment is compromised is easy and actually doable (as opposed to the case of trying to find malware on a compromised OS).
If, in addition, a user account has limited privileges (which it should of course), then even when compromised the chances of malware being able to do a lot are a lot less. For instance, turning off a webcam light being a root is probably easy, otherwise probably not. Setting up a server, listening for incoming connections and punching a hole in local firewall as root is available, but as a regular user is not.
Stop giving admin rights to your computer to random people (install software as an admin) and live much happier. As an additional benefit, there's never a situation 'I installed this and now computer is messed up, because Joe-the-dev ran "rm -f " with a wrong path as a parameter'.