The compose key is also useful outside of programming environments: I'm german but still prefer english keyboard layouts for easier access to several characters (like {}[]\ and so on) which require the use of Alt Gr on german keyboards. The compose key still allows me to use german umlauts like ä by typing <Compose>"a. I've gotten quite fast in doing that because by having compose on Capslock, I can use two fingers of my left hand to press the Shift and Capslock keys (nearly) simultaneously, one finger of my right hand to press the ' key (which, together with Shift, gives me the ") and then press a afterwards.
I don't think this is a problem with GitHub, it is - if you want to call it a problem at all - a problem with git itself because noone stops you from using any email adress and name as user.email and user.name in your git config. You could do the same thing on bitbucket, your companys git repositories, git repositories on your own server etc.
That's really not a solution to the original problem, just a short-term solution for this one conference. From the mail by Mark Messersmith it looks like the ban applies to all conferences held at a NASA building, so if the NASA ever wants to have chinese participants at one of them they'd have to hold it outside of their own buildings.
You really should not have to find people working at GitHub on other parts of the internet for something like dealing with copyright issues if you're not the copyright owner because (I suspect) copyright owners will have no trouble getting GitHub to (at least) listen to their complaints.
That was my thought as well after seeing the list of Free APC bugs. I can see why they'd want to limit the number of bugs that you can get free hardware for (to avoid trivial pull requests like fixing spelling mistakes etc.), but there should be a few bugs between the low hanging fruit and the ones you need the actual hardware for.
(I haven't gotten Trine 2 before, so this may not be correct) The HIB9 includes a Trine 2 Complete Edition ebook, maybe that's what makes the 'Complete Story'?
Having reported a bug in Chromium myself that I thought was security related (but in the end was not - better safe than sorry), I think it's a reasonable decision to not show security related bug reports/tickets from the general public - you'd otherwise present all security holes in the browser to every evil guy in the world on a silver platter.
That's actually how the german DHL website works - unless you enter the recipients postcode, all you get is information that's not telling you who the recipient is (that is, the shipment progress, the latest status). If you enter the postcode, the recipients name and the name of the person the package was delivered to (for example your neighbor) will be revealed.
If I understood the intention correctly, it's not about saving time downloading files when saving them from the internet to your computer or Dropbox, it's really about saving the time you spend uploading large files that you downloaded from the internet from your computer to your Dropbox. At least where I live the upload speeds on normal internet connections are way smaller than the download speeds, so what this app basically does (I think) is download the file to your Dropbox via their server, so you don't have to spend the time uploading the huge file to Dropbox because those guys did that part for you.
This is exactly what I feel like every time I see an article/presentation about (maybe even really small) tools that just get the job done but I didn't know about and didn't even think about looking for although I can think of so many cases where they would've been incredibly useful.
Imho GitHubs pull requests shouldn't be used as a replacement for good commit messages or code comments - GitHub might go away, you might decide you don't like them anymore or whatnot. If you feel like someone raised a question about some part of your code in a PR that does have an answer that's only obvious to you, there's no harm in just adding another commit adding comments to the code (you can even squash that commit into the commit changing the code between merging to master and pushing).
If you don't trust your package manager & distribution supplied packages (read: your whole operating system), the security risk between an 'apt-get' and 'go get' is most likely not the most important thing to worry about.
I really hope they finally fix the MD5 signatures on the Psychonauts FLAC soundtracks. The game and soundtracks were already included in the HIB4 with mismatching signatures and they haven't fixed it yet :(
I love how nobody at GitHub considered the possibility that a project might not use GitHubs issue tracker - any project not using it will still have the stats displayed with links to the open/closed issue that result in an immediate 404.
In that case you'd need to either force the next push because squashing rewrites history or use a separate branch to do the renames in the first place.
This is not an oversight, it's even mentioned in their privacy policy [0], although I also stopped at that point. Whenever I see a service requesting permission to even have a look at my contact list, I really wish there were a button like "No, I really don't need this social stuff because I know noone in my contact list uses your service anyway and if someone does, I really don't care".
I'm not a Debian user so I'd appreciate if anyone could shed some light on what's so wrong with Valves installer (it can't be that bad since there now seem to be packages available [0]) that made made it necessary to rewrite it instead of fixing it like the Debian guys seemed to do?
Don't get me wrong, I'm all for fixing stuff if you know how to do it but the first thing to do should always be trying to get it fixed upstream and/or by your distributions packagers.
[0] http://ftp-master.debian.org/new/steam_1.0.0.33-1.html
This article doesn't seem to contain any surprising information - yes, if you build a product on an API that's not under your control you should make sure you're always up-to-date with information about that API.
What I don't understand is why one would use yet another service (ways 3-5 in the article) to be notified about API changes. It basically means you're now relying on an API that might be changed at any point in time (it probably won't be without prior notification) without you having any say in it AND you now also rely on another service (that does not make a guarantee about a timely manner of notifying you) to get information about said API changes.
At the TU Ilmenau we used OMNeT++ (http://www.omnetpp.org/) which had the benefit of having built-in classes for statistics collection and looking at graphs while the simulation ran - mininet doesn't seem to offer such functionality (yet?). But then it sometimes felt like omnet had too many layers of indirection and C++ was not everybodys language of choice.
Our assignments were similar, I remember doing stuff with link state and distance vector routing, implementing parts of a tcp state machine and others.
> - For multi monitor support I had to install nvidia's drivers from the repo and set up the dual monitor config from their utility. (This is a PITA utility and I blame nvidia)
The proprietary driver has support for xrandr 1.2/1.3 since version 302.07 which has been released in last may, so you're no longer confined to using nvidia-settings (in case of Lubuntu probably lxrandr). This also solves many issues with let's-go-fullscreen-over-all-your-monitors applications.