hmm I guess when I said crappy code I wasn't clear about my lack of good programing techniques ...
and the paper clearly outlined that some people think its a hard thankless problem so maybe I should find something fun to do instead

Long read, confusing/obtuse notation with C* and "C" used to denote C optimized for an single platform and "ideal" "portable" C respectively. However Interesting in the end. Not sure it will make me avoid -O2/ -O3 flags or switch compilers, but I will keep this subtle C* "C" distinction in mind. Thank god I don't usually have to rely on "risky" behavior to get my crappy code to work.

if you hate this crap host your own

if your frickin lazy grab a premade VM from https://www.perforce.com/downloads/helix-gitswarm that has a Gitlab fork pre configured.

sweet how long until it some one forks something like notepad#

also everyone suggesting vi/vim don't forget to tell op to first install a virtual machine so he has access to it via putty ... ideally he should force IT to give him an old machine that he can hide in some odd closet or roof so he can putty in to edit his text to copy back to his crappy edition of VS2008 express the company is too cheap to pay for

one of these days I'm just going to go off the rails and pull an office space at my initech.

I'm really just waiting for when the bobs show up

well if you want security you have to accept that it may not be simple

if you want to change your blessed source you just need to point your copy to the source pull the changes there and resolve your copies differences against theirs prior to merging your work back in (something like this helps explain the process better than I can http://think-like-a-git.net/sections/graph-theory.html)

if you want simpler you will need to accept that you will need a single source of truth and that has its own risks

Um what makes it centralized in any sense of the word? The choice to Bless a one instance of the repository (usually server) as the source of truth is usually an arbitrary choice.

Git makes a pristine copy of the source to you system, it verifies that this copy is the same as the original source location using cryptographic hashes of the data, (it may not prevent you from losing data but you will know if it is tampered with) it was designed for decentralized coding with an emphasis of ensuring no one slipped in changes by accident or by nefarious design

if hosting it becomes illegal burn it to CD an mail it out, hell turn it into 2D barcodes and fax it to your contacts in a prearranged order that would be hard for others to figure out

fact is the contents of a project are is just data and everyone has the data if they have a copy of repository, any attack on how to share the data is an attack on communications architecture and that has as many work arounds as there are inventive people in the world e.g. https://en.wikipedia.org/wiki/IP_over_Avian_Carriers

so whats the source of this fearmongering ? is it someones inability to use the tool? or it it a confusion that Gihub = GIT? (spoiler Git hub is just one way among thousands to use git)

wonder if there is some crazy lawyer out there that can spin all the GPL 2.0 stuff to make it so you can't restrict free speech based on crazy societal norms

ah if only lawyers weren't selfish bastards

Matlab really is an odd duck. My feeling/experience is Simulink (the graphical part of Matlab) really jives with systems engineer and control systems guys.

Throw in the fact that you can turn a crank(a very expensive one called embedded coder ) and you can generate c/c++ that should run on practically anything that controls/systems engineers want to run on.

That said if you are doing image processing outside of matlab's set of expensive image processing packages your going to have a bad time

The place i work at is a 50/50 split right now,and as a guy its not that bad of an environment.

That said what we do is just outside of coding (more simulations and control systems) so maybe that helps keep it a more even split

Sometimes belonging to the same set as http://m.youtube.com/watch?v=71F_f15FXnk

I like to use cmake and eclipse, some times vim when its a remote system , cmake lets me use eclipse for a reasonably easy to use debugger (cause fuck learning gdb). I also like the fact i can bypass eclipses slow ass build environment and use make directly from the cmd line

Cmake also lets others use another ide if they have half a brain

im sure this beats it hands down

"Then I realized, it’s Steve Bromley from Conche … I used to buy my codfish from him" source--> http://news.nationalpost.com/news/canada/newfoundland-man-drives-18km-in-wrecked-car-cannot-remember-hitting-moose-hey-buddy-you-got-no-roof-on-your-car

sorry its national post but it is still funny

just say sorry eh. and offer them a molson or a labatts and move on. if they continue to bug you about it say quit acting like a quebecker

true,the VC redistributable may be different, however, i would be interested to see how much using something like POSIX Threads for Windows would suck in comparison to the native version. that said there would be no real easy way to do this sort of test.

i know its completely out of scope but it would be interesting to see how windows wrappers of C/POSIX looked in comparison

you don't deserve the upvote but using VSS in 2015 made me laugh