Hacker Newsnew | past | comments | ask | show | jobs | submit | benvd's commentslogin

>I don't use !!, or !$, because I prefer to see what command is being executed before I press enter.

I use zsh (with oh-my-zsh) and when I type e.g. "sudo !!" and press enter, it doesn't execute the command, instead it expands the !!. Another enter will then actually execute it.

I believe it's default behavior for oh-my-zsh. At least, I don't recall manually changing this.


This is achieved in bash by shopt -s histverify

I don't use history expansions, though, so YMMV


In zsh you can type !!<tab> and it will auto-expand the line for you. I find histverify to be kind of annoying so I don't use it.


How so?


I think by default in vanilla zsh, will expand something like 'sudo !!' by hitting tab after the second bang.


You are correct.


Same behavior on zsh with prezto.


When it comes to grammar, the term hypercorrection is typically used. Seems like it could be applied here as well.

http://en.wikipedia.org/wiki/Hypercorrection


Most people couldn't stand on one foot? Really? Am I that out of touch with reality?

Not even being snarky.


It's not that people can't stand on one foot, it's that if you stand on this wheel with only one foot, then you'll fall sideways before your other foot has a chance to hit the ground. Well, that's my assumption, anyway. Get on a (normal) bike, ride along, then stand up on it so your weight is on the pedals through your feet. Come to a stop, then try and get off it, from standing, without supporting yourself with handlebars or seat.


I've had an opportunity to try this contraption (or something remarkably similar) and this problem is academic: it's very much unlike a bike, but you'll learn how to stop and dismount without falling off, very quickly (haven't really tried whether I could stop w/o dismount, as I've played with it for a few minutes at most). It's similar to unicycles in this respect (note how in the videos, users ride back and forth a little bit, instead of coming to a complete stop).


OK fair enough, I've never tried anything like it - was just explaining why it wouldn't be quite as simple as "can you stand on one leg".


man they got this girl riding around in circles on one foot, you can't do that?


If enough people never bother to change their PIN, it might still be worth it.


Yeah, but then you have to make the lookup anyway, to check if they did change it. You might as well just authenticate. Unless the original PIN always works, which is not secure.



I much prefer Meld (http://meldmerge.org/) when I want a graphical tool, but usually just edit the conflicts in vim.


Out of interest, do you have a good way to compare two parts of the same file using vimdiff without creating two files? Say a huge xml file with two nearly identical <parts></parts>?


Yank the two parts into two new buffers (:vsplit) and do :diffthis in each


I've actually never used a visual diff tool. Usually I just search the file for ">>>>>>>>>".

Maybe I should check it out.


As someone who moved from perforce to VS/TFS2010 at work, the first thing everyone at the office demanded back was p4merge. It's a very good tool.

Fortunately, with VS2012 the built in merge-tools has become much better and allows you to do inline editing in the basic compare/merge-view as well. Since VS2012, I can't say I don't miss p4merge, but I miss it a lot less.


I've been happy with BeyondCompare. Reasonably priced, does 3-way merges, configurable with many version control systems. Now with Linux & Windows versions (but not a OS X version yet)


If you are into this stuff, Araxis Merge is also worthy of consideration. It does a better job of char-by-char diffs than p4merge, and will recalculate the diffs after you manually edit the result.

The (old-ish) version I used had only 3 panes, though, making p4merge superior for 3-way merges. p4merge was easier to use from the keyboard, too. But for ordinary merges, I much preferred Araxis.


SystemRescueCd is a great one. I always carry a USB drive with it on my keychain.


It says "Inofficial ebuilds" under the Gentoo section.


You're afraid people you know will be able to associate a review of a mobile app with you?

Genuine question. I really cannot fathom why that would matter.


Why in the world would you want your review of a random mobile app publicly associated with you? I would rather just not bother posting a review.


Because it gives weight to your opinion.


Why do I care how seriously people take my review of a mobile app? I'm just posting it to help others make a decision about whether to buy/download the app. There's not much in it for me either way. If there is any potential downside at all, even a small one, I'm not going to post a review. I suppose some people care about being a top, respected reviewer, but not most.


Writing a quality review should be all that is needed to add weight to your opinion. Why does it matter if I know it came from Tom Jones in Skokie, IL?


Because a lot of times "Tom Jones in Skokie, IL" is actually a competitor going around giving everyone else a negative review


Can such a competitor not create a "Tom Jones in Skokie, IL" Google+ (along with a myriad of other fake accounts) to achieve the same result? Or, is that to difficult for someone with those intentions?

Edit: Also, "a lot" is a rather unhelpful metric.


My point is that a good review (not necessarily a "positive" review) is a good review whether I know who gave it or not. Is a detailed review from an anonymous user giving something 2 stars somehow worth less than a one-liner review from a "known" user giving it 5 stars?

That aside, as greyboy already points out, what is going to stop a competitor from creating a fake Google+ account to give that negative review? This move doesn't prevent that hypothetical from happening in any way.


Health applications - a review for an app which allows you to record you stool consistence (for Morbus Crohn) could negatively affect your chances for employment as well as open you to ridicule by others.

"You gave AcmeFamilyPlanner 4/5 stars, Mrs Candidate - say, do you want to have a career or a family?" (Hm, of course this could help remove bad employers from your list of prospective work places ;) )


  > "You gave AcmeFamilyPlanner 4/5 stars, Mrs Candidate -
  > say, do you want to have a career or a family?"
I am almost certain that questions like this are illegal in any nation a HN reader would want to work in.


You wouldn't ask that question, but use that information to disqualify the candidate before the interview.

Just because you can't legally discriminate based on certain things doesn't mean people don't find ways to do so anyway.


The point is that the employer wouldn't even have to ask the question. The data is now public.


Okay, thats one app. What about all the other innocuous apps you download?


Politics app. Sex app. Porn app. App about guns. Violent game app. Match.com or OKCupid app.

Pretend you're a school teacher with a few malicious students that have access to Google. This exposes to them what you have privately on your phone, so you're no longer going to rate apps in the ecosystem.


Personally I don't have, nor do I wish to have, a Google+ account.

No reviews from me.


I have one because a lot of tech people are on there and it's easy to get news I don't wanna subscribe to. But given that Google won't allow me to use my actual internet identity of ten years, I've merely got the account set up as a first name and an indistinguishable initial. Google have done a fantastic job in alienating me as a user. What's hilarious is that the name I want to use and wouldn't mind associating with my G+ reviews is the same name that I use on my email account. Google apparently doesn't care though, so no reviews from me either.


On the contrary, if it doesn't matter, why associate them?


So put your zsh config in a git repo, clone it wherever you need it and symlink to the repo?


Uh huh. Now I have to install git on all those computers as well as make sure there's a recent-enough copy of zsh. Much less work that way.

Or I can just use the bash that's already there.


Your concern about git however is unrelated to shell choice. I customize bash quite a bit (my own completions, etc.) and maintain dotfiles in a git repository. I have a post-receive hook that tars up my dotfiles and puts them at a public location on my http server that only I know about so that I can easily get them when I'm on a machine without git. I'm not going to do that for every machine, but if I'm stuck debugging something on some odd machine, at the very last I want my vim settings around.


It makes more sense for vim than zsh though. What color your prompt is has a lot less bearing on your productivity than your editor configuration. (Though I don't do much of that either).


I am of the same practice as fusiongyro. I also expand on that and have a very simple configuration for Vim. Once you go beyond a handful of machines, you never know what to expect on the system. "Lowest common denominator" wins out for me almost everytime.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: