News for hackers. Stuff that matters.

anonym

Re:The Costs of Bookmarking on Sunday September 11, 2011 @04:46PM (#2984781)
Don't you think web crawling (for users with archival accounts, where the full text + associated images/resources for each bookmark is stored and indexed on Pinboard's servers) is probably using up more of those resources than actually serving up pages? I don't think 10 pages per second or whatever is really the relevant metric for this app.
Re:Why Javascript Creeps Me Out on Thursday July 03, 2008 @02:27AM (#234963)
Yes.

I have to admit that this behavior confused me too when I was first learning Javascript. My only previous experience with lexical scope was in Common Lisp, which of course works the same way -- but the issue doesn't seem to come up as much there, perhaps because straight reassignment within the same lexical scope (of non-special variables) isn't something you do too often; perhaps because you can establish a new lexical scope and simultaneously rebind with LET within the normal control flow without defining and then calling a function, or perhaps simply for stylistic reasons). So I'd never really noticed it in Lisp. But once I realized what was going on it was clearly perfectly sensible. It's hardly "creepy," and it's certainly easy enough to establish new scope when you need to.

And for crying out loud, it's not got anything at all to do with asynchronicity.

Re:HTTP Cache Poisoning via Host Header Injection on Wednesday June 11, 2008 @09:28AM (#214829)
"<a href="http://evilsite.com#/about">About</a>"

There is absolutely no reason to build links with your own hostname in them in the first place. Why on earth would you make a link like this instead of href="/about"?

edited to add: the spec says that the Location header for a redirect must be an absolute URI, so that's one case where you'd need to know your hostname. But you certainly don't need it to build links.

Re:Bringing Sexy Back: 280Slides on Saturday June 07, 2008 @09:11PM (#211809)
It's certainly technically extremely impressive.

The user experience, though, is not great, and certainly not worth the effort it must have involved. Why reimplement draggable, modal windows? What does that get the user? Must we really duplicate all the worst excesses of GUI interfaces just because we can? There are some really brilliant moments here (e.g. the ability to add pictures directly from Google Image Search) where the designers clearly bothered to think about how exploit the strengths of web-app-ness, but a little more of that and a little less slavish devotion to the GUI paradigm would not have gone amiss; there's too much "look what we can do." Just because you can doesn't mean you should.

Also, the author of the linked post claims that it is responsive, which it certainly is not for me; it feels very, very sluggish, 1+ second delays for mouse clicks to register.

Re:Poll: Suggestions please, which JavaScript library? on Saturday June 07, 2008 @06:04PM (#211743)
I use Dean Edwards's base2. I like it because it is very simple, and because it adheres to standards; that is, rather than creating a map and a forEach of its own, with its own semantics, it implements them according to the Javascript standard (or proposals, don't recall if they have yet been adopted) for browsers that don't yet support them natively (anything but Mozilla, I think). It also cleans up DOM manipulation so that you can use standards-compliant methods and it will work on non-compliant browsers (all the way back to IE5).

Similarly, for DOM querying, it implements the W3 Selectors API rather than reinventing the wheel.

I really like this approach. base2 doesn't provide anything in the way of widgets (though I think there may be some code along these lines in the Subversion repository, implementing some of the HTML5 components) -- for most of what I do, widgets aren't available anyway or would need to be extended, so I end up writing my own. And it's easy enough to use a YUI or other widget with base2 when it makes more sense.

For animation, I've used animation.js and found it simple and fast.

Re:Ask HN: What web language for server side scripting? on Wednesday June 04, 2008 @03:45PM (#209304)
If you want speed, why not Common Lisp? Nice, high-level, a pleasure to work in, fast and compiled out of the box, with the ability to optimize the hell out of specific critical parts of your code (using, e.g., optional type declarations, speed and safety settings, inlining declarations, etc.). Hunchentoot is a nice web server, Postmodern is a nice library for working with Postgres.
Re:Lisp: No easy download on Sunday June 01, 2008 @03:02AM (#205719)
Why are you building, instead of downloading the binary?
Re:Lisp: No easy download on Friday May 30, 2008 @01:41PM (#204638)
http://www.sbcl.org -> "Downloads" -> Linux/x86 (or your architecture/OS of choice).

Hold on a second, I need to lie down for a moment after all that hard work there.

OK. Yes, there's not one Lisp implementation that is canonical and called "Lisp" and which lives at lisp.com. Life is pain.

Now that's just cold.
What's Common Lisp?

I'm thinking either Barry Goldwater or Eugene McCarthy, kind of a toss-up.

Re:Ask HN: good place to store code online? on Tuesday May 27, 2008 @01:14PM (#201188)
No, I backup my home directory and some web directories.
Re:Ask HN: good place to store code online? on Monday May 26, 2008 @11:05PM (#200637)
I use http://duplicity.nongnu.org/ to backup to Amazon S3 (it has support for various other backends). It encrypts with GnuPG. S3 charges a few cents per GB of transfer and a few cents per GB of storage per month.
Re:Web Applications are Overrated on Monday May 26, 2008 @10:54PM (#200632)
"We now have technologies that effectively mix web and desktop technologies, something I like to call 'webtop clients', manifest in Adobe Flash Player, Java Applets, and Microsoft Silverlight."

1. Flash is used for: video, music, ads (usually also only to enable video), and sites for the sort of web design shops you don't want to hire. Also, games, a special case. When was the last time you saw Flash used to recreate a desktop-GUI-style app? That OpenLaszlo (sp?) sure is taking the world by storm, eh?

2. Nobody uses Java applets.

3. Nobody uses Silverlight.

Statelessness is the core strength of the web, and the core strength of web applications. The limitations the web naturally imposes make for better applications and a better user experience than the GUI paradigm. (Every time I hear about some great new continuation-based web framework, there's always an example showing how easy it is to create the equivalent of a modal dialogue, and I always want to hit the writer over the head with a copy of a decent book on interface design. Modes are bad! Statelessness is good! The resource-centric model is good! Users like these things for the same reason you like REST better than SOAP and modular code rather than otherwise — people don't do modes!

(Sorry for the rant, these GUI partisans get me riled up.)

I am reminded of Borges's "Pierre Menard, Author of the Quixote" (http://www.coldbacon.com/writing/borges-quixote.html) (and what do you know, Borges even slips George Boole into it).

"He did not want to compose another Quixote —which is easy— but the Quixote itself. Needless to say, he never contemplated a mechanical transcription of the original; he did not propose to copy it. His admirable intention was to produce a few pages which would coincide—word for word and line for line—with those of Miguel de Cervantes."

Re:San Jose is hell on earth on Tuesday May 20, 2008 @10:13AM (#194931)
And calling a place nicer than a shopping mall is like saying that my squalid, decomposing first apartment was nicer than a prison cell; true, and irrelevant.
Jesus Christ, I just watched a video of Ken Tilton ranting on a beach (!?) and agreed with almost everything he said. That can't be a good sign at all.

(Hey smanek, this is eli)

Re:10 Most Beautiful Looking Linux Desktops on Sunday May 18, 2008 @04:37PM (#193595)
I don't really know enough about Linux fonts to answer that. The window manager, Ion, does not support Xft because its author hates anti-aliasing, so I'm not sure what is rendering the title bars -- just X, I think. The terminal font is Inconsolata as rendered by rxvt-unicode with Xft support. (Emacs is being run -nw, so it's just using the same terminal fonts.)
Re:10 Most Beautiful Looking Linux Desktops on Sunday May 18, 2008 @04:17PM (#193577)
Nice. I like flat desktops; I will never understand the appeal of faux-3D interface elements.

I must, however, confess to a weakness for transparent terminals.

http://scratch.flyoverblues.com/screen.png

Re:Real-World Tufte Graphics in 11 Lines of Code on Friday May 16, 2008 @05:44PM (#192252)
Vecto (http://www.xach.com/lisp/vecto/) in Common Lisp is excellent for this sort of thing (see http://common-lisp.net/project/adw-charting/, which uses it).
Re:How to Live Well Without a Job and with (almost) no Money on Wednesday May 14, 2008 @01:09PM (#189774)
They call 'em session beers for a reason.

(He says, surrounded by empty cans of Old Style ($5.99 the 12-pack) and the husks of half-written late-night defuns.)

Re:Why Yelp Works on Tuesday May 13, 2008 @11:57AM (#188678)
"It didn’t try to pay for reviews, as some sites have."

It certainly did, as someone points out in the comments at the Times. Craigslist was littered with ads for Yelp reviewers a few years ago when I was last looking for a job.

Re:Ask YC: What should I ask John McCarthy? on Tuesday May 06, 2008 @02:43PM (#182689)
Lisp is not a functional language except in the trivial sense that it supports HOFs. Common Lisp is not even functional in the sense of Scheme, where purely-functional programming is not enforced but seems to be encouraged (disclaimer: IANAS). Common Lisp is a pragmatic, multiparadigm language in which functional, imperative and other styles can and should be used as appropriate. In a very real sense Lisp is that language "in the middle."
Re:Why I hate Flash on Thursday April 17, 2008 @12:12AM (#166023)
"You don't have to read past that first sentence to realize that this is heavily biased opinion from a person who earns his living doing web design."

Absolutely! Really, who could be less qualified to comment on web design?

Re:The Typography of Code on Wednesday February 06, 2008 @05:19PM (#110496)
I've started using Inconsolata, which is both free and pretty nice. (It has particularly nice parentheses, which I appreciate as a Lisp programmer.)
Re:MapReduce: A major step backwards on Friday January 18, 2008 @08:58AM (#100303)
[Note: Although the system attributes this post to a single author, it was written by David J. DeWitt and Michael Stonebraker]

Oh, the irony.