News for hackers. Stuff that matters.

wonderercat

Re:Just start an experiment, not a damn startup on Thursday March 22, 2012 @12:58PM (#3740698)
> Set a deadline of 2-4 weeks to ship the first version of the product.

I blame this kind of mentality for much of the derivative rubbish the startup ecosystem is polluted with. We don't need any more photo/video sharing, link voting, or file sending apps.

I'm all for validating assumptions early, but some ideas (the ones that matter) are much, much bigger than 4 weeks to an MVP.

Incidentally,

http://paulgraham.com/ambitious.html

What amazes me is how Microsoft consistently explains design with obviously aliased images using nasty, bandy palettes.

I hope the explanation is that their web publishing framework just sucks and has no oversight, because it would be troubling if the designers/bloggers missed this.

Re:$100 to Fly Through the Airport on Thursday March 15, 2012 @01:22AM (#3707001)
> Create a problem.

> Sell a solution to that problem.

Force everyone in the country, by law, to buy the solution.

I'm not even talking about the latest TSA BS; are we all forgetting that our taxes -- our money -- has been literally purchasing the TSA's nonsense from the very beginning?

Re:Show HN: Stack Overflow meets MTurk - Minute Favors on Wednesday March 14, 2012 @05:20PM (#3705422)
...which is easily defeated by a regex, no?
Re:Show HN: Stack Overflow meets MTurk - Minute Favors on Wednesday March 14, 2012 @05:13PM (#3705393)
I'm doing a (tangentially related) twist on this sort of thing, so I'll ask you the questions that haunted me for months until I could answer them about my own project:

Who's going to use this?

I mean, what's the intersection of people that are computer-articulate enough to know what to ask for ("convert this file for me"), but don't know how to do it, and/or are willing to spend the extra minute or two to post and manage the "favor"? Especially when "know how to do it" means "google for the free web app that does it"?

And who is going to spend their time doing these favors, with no reward except some ethereal concept of "points" that can only be redeemed for doing the things I can already easily do myself (see above)? Realize that you're competing for attention from facebook/twitter/HN. And MTurk, which actually pays.

Re:Show HN: Password validity hinting on Saturday March 10, 2012 @06:00PM (#3688965)
You could still use bcrypt and do the comparison work server-side.

Depending on work factors and whatnot, it's now a backend load balancing issue. But it's probably solvable.

Re:Show HN: Password validity hinting on Saturday March 10, 2012 @05:49PM (#3688937)
I suspect they are matching bits.

Though since this is going through hashing, the number of matching bits has (by design) zero correlation to how close the P/W is. In which case, yeah, I think the binary indicator would be more useful and intuitive.

Re:In Which I "Strongly Caution" The TSA To Snort My Taint on Friday March 09, 2012 @02:42PM (#3685605)
> My guess is the machines could be modified to change the background color behind the scanned person's silhouette.

For a billion dollars, I would have demanded, at minimum, that modifications to make the machines actually work be included. That they weren't just underscores the idiocy at play.

> The quote from the TSA ("Any guidance provided is to caution reporters not to generalize...") is merely saying "don't jump to conclusions on the basis of some activist's blog post/video," which is quite a reasonable "cautionary" statement.

It would be reasonable, if this were a citizen petitioning the media. Except this is the government, funded by the people, telling the media on their behalf what should and should not be discussed. The most charitable analysis is that my hard-earned tax dollars are being flushed on unsolicited media curation. A less charitable (and I believe, more likely) explanation is an attempt to sidestep the first amendment with no repercussions because it's not censorship, just chilling effects which nobody can be held accountable for.

So at best, complete incompetence, and at worst, conspiracy. Either way, I think the outrage is justified.

Re:A fresh new Dropbox on the web on Friday March 09, 2012 @05:22AM (#3683586)
About as much of a stretch as using a unicorn for "delete", because unicorns don't exist.

I'm getting increasingly concerned that web UIs are trading away usability for design in Apple's wake, forgetting the part about "design is how it works".

Re:Microsoft's Azure cloud down and out for 8 hours on Wednesday February 29, 2012 @10:47AM (#3648218)
I think it's interesting that the #1 feature on their features page is labeled "Always up. Always on."
Re:Microsoft's Azure cloud down and out for 8 hours on Wednesday February 29, 2012 @10:35AM (#3648180)
And unfortunately that with outages like this, that number's probably not increasing anytime soon.
Re:Mercurial vs Git: Why Mercurial? on Friday February 24, 2012 @03:50AM (#3628496)
Once again you're demonstrating that you really don't know how to use Mercurial, but feel the need to bash its usability anyway.

   hg update <branchname>
will do exactly what you want. It's even the same command as in SVN.
Re:Maddox - I hope SOPA passes on Thursday January 19, 2012 @07:23AM (#3484183)
> There are a number of publishers on this list, including my own publisher. If the consensus I get from readers is that we should boycott publishers, I'll support the boycott even though it hurts me.

As much of an asshole as he is, I've always respected the guy for his shocking lack of hypocrisy.

If movie producers had the decency to make this kind of public stand, things might be different.

Re:This game was made entirely in css (no javascript) on Thursday January 19, 2012 @05:14AM (#3483865)
Most likely because of things like

  @-webkit-keyframes gameover{
Re:This game was made entirely in css (no javascript) on Thursday January 19, 2012 @05:12AM (#3483861)
Indeed. I had to audit the source believe it, but this is actually legitimate.

Massive abuse of data url's and keyframes, but it works.

Re:The Daily WTF goes white to "support" SOPA on Wednesday January 18, 2012 @04:05PM (#3481838)
> I remember when one of the pro-SOPA speakers countered an argument against the feasibility of this legislation with "Google's got the technology."

If I remember correctly, the argument was that since Google is effective at taking down child pornography, detecting something as simple copyright infringement should be a breeze.

When a Google spokesperson explained why that made absolutely no sense given that often not even the copyright holder can tell what's infringement and what's not, legislators scoffed, ignored what was said, summarized it as "we have the technology", and for lack of time pushed to move to other issues so that the bill could be passed ASAP.

> I'm pretty sure that if you just "connect the uploader's POST to the downloader's GET" then the Node server would have to buffer the entire file from the submitter before sending it to the clients.

Not at all. Use stream.pipe(). I've done it countless times and it works really well.

> The prospect of being bankrupted by a civil suit or thrown in prison is enough to stop most people.

Yes, but... if you're good enough to be hired by these companies, leaking source code without being traced isn't going to be hard for you. I've thought about this before and I'm actually surprised it doesn't happen more often.

Re:DNS provision pulled from SOPA, victory for opponents on Friday January 13, 2012 @11:12PM (#3463437)
Better to play whack-a-mole than play Manhattan Project.

And nobody has to play whack-a-mole. iTunes is a perfect example of "getting with the times".

Re:Microsoft confirms UEFI fears, locks down ARM devices on Thursday January 12, 2012 @10:48PM (#3459105)
For the purposes of this argument, they do. Apple is the client; they buy the manufacturing.

Whereas with Microsoft, the factories are the clients of Microsoft OS's.

Re:Poll: Do you think HN should go dark in protest of SOPA? on Thursday January 12, 2012 @12:14AM (#3455134)
Ironically, the reason this won't work is the same reason SOPA won't: proxies. HN is one of the largest gatherings of hackers anywhere; give us a break.

That's in addition to the fact that it's consensus that emails aren't worth a damn here.

And also the fact that it would take significant wasted effort to implement.

If PG did something like this, I'd lose all respect.

Re:Inner-platform effect on Monday January 09, 2012 @11:48AM (#3443167)
While I totally agree that the subject of the article is a Thing (read: "notable"), this is an inflamatory essay and complaints forum more than an encyclopedia entry in its current state.
Re:The Invention of Waterloo (Canada’s Technology Triangle) on Sunday January 08, 2012 @05:16PM (#3440683)
As a UW alumnus doing a full-time startup a few blocks from UW, I actually don't see much of this "vibrant start-up scene" happening.

This is entirely my fault, which leads me to the question I've been wondering about for a while now:

Where in town would I go to get hooked in?

VeloCity has always been on my radar, but their site is terribly uninformative and doesn't give much of a place to start.

I hang around the university quite a bit but this hasn't gotten me any leads, either.

Re:IPhone 4 in pure CSS3 on Tuesday January 03, 2012 @02:53AM (#3418969)
"Demos like these are cool if they are well done, and correct. Otherwise they feel more like a gimmick."

It's pretty obviously a gimmick (and a cool one at that).

Looking for pixel perfect accuracy is completely missing the point.

I took a quick look and here's what I've gathered.

This thing gives the impression that user code is pre-emptible (it really just forks the process under the hood), whereas using vm naively risks blocking your process (infinite loop in user code will gimp the server).

There's also some pseudo-console.log hack to allow communication from user code to the outside world.

Even so, I don't see the point of this either. vm + forked child process will do the same thing, it's much more versatile, and if you're working a use-case where running untrusted code is necessary, you're probably good enough with Node to cook up this kind of solution in no more than a few minutes.