hartez · HackerTrans

hartez

no profile record

Submissions

A proposed standard for source control robots

github.com

1 pointsby hartezvor 14 Jahren1 comments

comments

vor 10 Jahrendiscuss

> Why would we do something that benefits far away humans over something that benefits nearby humans? Don't help those foreigners (or people in NY)! Build a wall! Trump 2016! (Same idea, just directed at a slightly different target.)

You've misunderstood his sentence. He's not saying that we should prefer fixing things locally to fixing things globally. He's asking why we should believe that the tech elite can improve life on a global scale when they can't even improve it on a local scale.

vor 10 Jahrendiscuss

That actually sounds like a fun Roslyn analyzer project.

vor 10 Jahrendiscuss

Why would they do that?

vor 10 Jahrendiscuss

I honestly thought you were being ironic at first. The pricing info is in their FAQ: https://slimfaq.com/faq (specifically at https://slimfaq.com/faq/24-getting-started/84-how-much-does-...)

vor 11 Jahrendiscuss

The default terminal got a bit better in Windows 10. I've been trying to run stock for the last few weeks (rather than installing ConEmu (https://conemu.github.io/), which is what I used on Windows 7) and it's been mostly okay. They've still got a long way to go, but at least Copy and Paste finally works.

vor 13 Jahrendiscuss

And technically you should just use the 'ProgramFiles' environment variable, since the folder might be called something else if your install is in another language.

And Windows allows side-by-side installs of 64- and 32-bit versions of the same software; so installers should be aware of 'C:\Program Files (x86)' and 'C:\Program Files'; the first is for 32-bit programs on 64-bit machines.

vor 14 Jahrendiscuss

I'd prefer opt-in, and a nice directory of robots you could essentially subscribe to. I did a little thinking about this a while ago when the Gunio robot was annoying some people about whitespace:

https://github.com/hartez/vcs-robots

vor 14 Jahrendiscuss

And all that shuffling costs tax dollars.

vor 14 Jahrendiscuss

My wife and I opt out every time. Whenever possible, we just aim toward the line that's using the old metal detectors* (since the new machines are either slow or prone to breaking, this is an option most of the time). But when the only option is the new machines (backscatter or mw), we opt out.

FWIW, we have a friend who sells mw equipment, and he refuses to go through them for safety reasons. So even if I weren't opting out for moral reasons, I wouldn't go through them.

* I'm sure glad no terrorists are smart enough to do this.

vor 14 Jahrendiscuss

Can't watch the video right now, but the summary at the bottom mentions France and Bulgaria - does it bother to mention the geographic differences between the US and those countries? I feel like a broken record every time this discussion comes up, but no comparison is complete without addressing the differences in geographic distance and population density that US providers are dealing with.

vor 14 Jahrendiscuss

Yeah, I still don't understand the logic behind this. I've used the App Store and the Windows Phone Marketplace, and of the two I've actually bought a lot more stuff from the Marketplace. The concept of "trial" software where you can later upgrade to the full version if you want is built right into the Marketplace and the phone OS, so it's fairly easy for developers to offer it if they want. When confronted with several apps that do essentially the same thing, I pick up the trial version of each one and then buy the full version of the one that works best. I've bought a lot of apps this way, almost all of them more than $0.99. If I decide to try a new version of a different app later, I can do so with pretty much no risk.

Does anyone know why Apple doesn't offer this? Are they just worried it will somehow cause customer confusion?

vor 14 Jahrendiscuss

In this case it's not as much about keeping it secret (since anyone who sees/handles it might have the number memorized or recorded) as it is reducing the attack surface. If the full number is on a discarded receipt (or on any combination of two receipts) along with my signature, anyone with access to my trash now has a big chunk of the info they need for identity theft.

I'm not saying that obfuscating the first 12 digits on a receipt solves the problem; just that it's a very minor adjustment that makes things more difficult for the attacker. But some organizations are still failing at even these most rudimentary steps.

vor 14 Jahrendiscuss

Sadly, I can :)

Just a few years ago I looked at my receipt from a local Dairy Queen and it had all but the last four digits printed on it. I complained to the guy behind the counter, who didn't see the problem.

I assume most new POS systems nowadays don't even have the option to print anything but the last four, but there are still some out there in the wild which do. Which is why I still shred all my receipts.

vor 14 Jahrendiscuss

That's what the last three paragraphs of the post are: suggestions for solutions to the problem.

vor 14 Jahrendiscuss

I think they meant that no one except for McDonald's was allowed to serve french fries (or anything similar to a french fry).

vor 14 Jahrendiscuss

just try managing AD users in PowerShell - sure its easy once you've written your own custom script or mapped it to a web interface or something, but there's no built in equivalent for adduser or about a billion other basic UNIX commands

It's not that bad. Adding a new AD user is a built-in command. From the docs:

New-ADUser GlenJohn -OtherAttributes @{title="director";mail="[email protected]"}

I'm not a UNIX admin; maybe the adduser command is easier. But managing AD stuff in PS is pretty easy (especially if you have to do it in bulk).

vor 14 Jahrendiscuss

Ran into a similar one about ten years ago; a medical records system with the assumption that you wouldn't need dates before 1/1/1900.

You know who goes to the doctor a lot? Centegenarians.

vor 14 Jahrendiscuss

From http://msdn.microsoft.com/en-us/windows/hardware/hh852363.as... (under the section "Updated or Removed Features"):

      The Windows SDK no longer ships with a complete command-line build environment. The Windows SDK now requires a compiler and build environment to be installed separately.

Below that they also list a dozen tools and a bunch of documentation and samples that they also no longer include in the SDK. My guess is that they're just trimming down their SDK downloads. Their SDK has been getting more and more bloated over the years with utilities and samples and stuff that most devs don't need.

Completely dropping their command line tools doesn't make sense for anyone who runs build servers; it doesn't make sense for their Powershell tools; and it really doesn't make sense when you consider their recent moves in Open Source. My guess is that the command line tools will be separate downloads from the basic SDK and you'll still be able to write C# code in Notepad++ if you like. The basic SDK will be targeting people writing Metro apps (which they want to encourage by default), but they aren't going to just drop everything else. This is a company that lives and breathes backward compatibility, after all.

It's just a case of poor messaging by Microsoft. Remember last year when everyone thought that all Windows 8 apps were going to be Javascript/HTML 5?

vor 14 Jahrendiscuss

The net result is code which breaks in spectacular fashion when the first unexpected condition is met.

Which is generally what you want to have happen. Failing fast is a good thing. If your code does anything other than "stop executing" after encountering an exceptional situation, it should be very something carefully considered and well-thought-out. With checked exceptions, the "I'm in a hurry to get this working" version of the code often ends up looking a lot like:

  try {...}
  catch(ExceptionType ex){ //TODO figure out what to do here }
  ... keep going ...

The danger is what happens in that "... keep going ..." section - the code is working under faulty assumptions and that can lead to a lot of non-obvious but horrific bugs. In the unchecked "I'm in a hurry" version, the program simply crashes at that point.

So I think the C# design choice leads to the least pain under bad conditions. I'd rather have a fragile application than one just robust enough to write bad data to my database.

That said, it's not perfect. Any sufficiently lazy/stupid programmer can write the equivalent of "On Error Resume Next" in any language.

vor 14 Jahrendiscuss

"People used to say shit like this on slashdot all the time"

I think that to myself about at least one HN article/comment every day :)