Hacker Newsnew | past | comments | ask | show | jobs | submit | Nicolay77's commentslogin

I think that Vista should be made immune to piracy hacks.

The reason? It would encourage more people to try alternatives.


Opera has been doing that for ages.

AFAIK Opera 9.5 Beta is still faster than these two browsers.


Where did you come up with that assumption? A quick (very unscientific) test of a big web application (over 100 JavaScript and image files) shows you're wrong...

(first column is seconds with empty cache, second column is seconds after cache was filled)

    WebKit     - 15 / 9
    Firefox 3  - 18 / 12
    Safari 3   - 19 / 12
    Opera 9.24 - 20 / 15
    Opera 9.5b - 20 / 19
    Firefox 2  - 21 / 21
Again, these are very rough numbers (I took the best out of about 3 runs) on a pretty slow machine (1.33GHz G4). Perhaps the numbers would be different for simple static pages, but I don't have a good way to time that.


I got my data from here: http://celtickane.com/webdesign/jsspeed2007.php

Of course, the scientific method means this tests should be reproducible by everyone, and YMMV, but I wasn't talking bullshit, just see the page.


Separating interfaces from implementations is ALWAYS good design.

If I understand correctly, a pure virtual interface avoids inline functions. And there can be a lot of one-liner inline functions in a large project.

So there's a diference between standard good-design and this yet-another-thing-that-became-a-pattern.


Hype.


Hehehe, I do use and administer a Subversion server, but to me the webdav choice was ugly.

Standalone svn:// server for the win.


Webdav is really one of the really ugly design decisions of the Subversion project. WebDAV was initially promoted by Rational - if you closely study WebDAV, it is easy to notice similarities with Clearcase. Looks like it was a political choice that was made to appeal to the managers in the IT departments - "Oh, Firewalls are not a problem, If your browser works, then the SVN should work ... blah blah". To be fair though some of the top committers will freely confess to a dislike for WebDAV.


Or svn+ssh://


I always use svn+ssh. I had it under HTTPS initially (first using Apache, then Lighttpd), but it had a strong tendency to corrupt the working copy. Like daily. So I had to go in and delete .wcprops after every substantive check-in, which got really old really fast.

No problems with svn+ssh; seems to work fine.


He's just trying to impress the lady in black.

(Or some other lady nearby.)


AFAIK the $20 firmware upgrade exists because tax laws, not because Apple likes it that way.

Proof: it was different before, and Apple got in legal trouble.


I think that compilers should be specially easy to write in Lisp over any other language.

However in the time that most people were writing compilers no common or popular computer had enough memory to run Lisp reasonably. Even 8MB were a luxury.

Now we have 2GB of RAM as the norm.


Almost all 'design patterns' are just instances of yet another pattern: The Human Compiler.

Make your language do something it can't naturally do, by having to transform abstractions into hand crafted code. Especially when using a higher level language would make that 'design pattern' automatic or innecesary.

That said, I'm not agains patterns per se, I just don't see them as the 'Holy grail' that fixes all the developer problems if only you used them well, because they are nothing of that sort.

'Design patterns' are simply like an Assembler programmer doing ADTs, just because he knows about OOP, and he is methodic about his work, but it doesn't mean that what he's doing is the only way to use classes when programming.


Funny, I started doing a lisp interpreter a while back and I have only got to the part where it makes a damn fine calculator.

I believe my version is much handier than having to use Emacs.


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

Search: