Comments for hatsuseno's blog http://blog.hatsuseno.org rantings on tech Sat, 05 Sep 2009 15:42:33 +0200 http://wordpress.org/?v=2.8.4 hourly 1 Comment on gnome-terminal gconfd bullshit by hatsuseno http://blog.hatsuseno.org/index.php/gnome-terminal-gconfd-bullshit/comment-page-1/#comment-448 hatsuseno Sat, 05 Sep 2009 15:42:33 +0000 http://blog.hatsuseno.org/?p=12#comment-448 And so I have. I won't be able to tell the difference between roxterm and xfce's terminal, it's all VTE. But it's good to have the configuration files for all three major GTK/VTE terminals now, this way I won't have any problem switching between boxes that I don't have root on. And so I have. I won’t be able to tell the difference between roxterm and xfce’s terminal, it’s all VTE. But it’s good to have the configuration files for all three major GTK/VTE terminals now, this way I won’t have any problem switching between boxes that I don’t have root on.

]]> Comment on gnome-terminal gconfd bullshit by maze http://blog.hatsuseno.org/index.php/gnome-terminal-gconfd-bullshit/comment-page-1/#comment-443 maze Fri, 04 Sep 2009 18:52:13 +0000 http://blog.hatsuseno.org/?p=12#comment-443 I'm in love with xfce4-terminal, it's clean, elegant, borderless, scrollbarless, translucent and has great font rendering... you should give it a try :-) I’m in love with xfce4-terminal, it’s clean, elegant, borderless, scrollbarless, translucent and has great font rendering… you should give it a try :-)

]]> Comment on gnome-terminal gconfd bullshit by Hans Otto http://blog.hatsuseno.org/index.php/gnome-terminal-gconfd-bullshit/comment-page-1/#comment-7 Hans Otto Sat, 13 Jun 2009 17:22:18 +0000 http://blog.hatsuseno.org/?p=12#comment-7 I had same problem, but for me the solution was to run "gconftool-2 --spawn" explicitly before running "gnome-terminal". According to the documentation of gconf it should start automatically, whenever needed. But here it didn't. So I just added "gconftool-2 --spawn" to my .xsession. I had same problem, but for me the solution was to run “gconftool-2 –spawn” explicitly before running “gnome-terminal”. According to the documentation of gconf it should start automatically, whenever needed. But here it didn’t. So I just added “gconftool-2 –spawn” to my .xsession.

]]> Comment on PHP and XML manipulation by hatsuseno http://blog.hatsuseno.org/index.php/php-and-xml-manipulation/comment-page-1/#comment-4 hatsuseno Fri, 29 May 2009 00:03:49 +0000 http://blog.hatsuseno.org/?p=32#comment-4 Hmm, you did provide me with an idea, using PHPs Reflection API maybe I could adjust the read-only behaviour in run time, but sadly, it seems that as with SimpleXML, DOM objects are constructed with some kind of black magic that prevents explicit reverse engineering on the classes. There are no properties, static or otherwise, only the methods already described on php.net. Would have been nice though :) Hmm, you did provide me with an idea, using PHPs Reflection API maybe I could adjust the read-only behaviour in run time, but sadly, it seems that as with SimpleXML, DOM objects are constructed with some kind of black magic that prevents explicit reverse engineering on the classes. There are no properties, static or otherwise, only the methods already described on php.net. Would have been nice though :)

]]> Comment on PHP and XML manipulation by Juerd Waalboer http://blog.hatsuseno.org/index.php/php-and-xml-manipulation/comment-page-1/#comment-3 Juerd Waalboer Thu, 28 May 2009 23:47:53 +0000 http://blog.hatsuseno.org/?p=32#comment-3 Immutable (read-only) data is annoying in many ways, in many libraries and frameworks. The main problem is that once you've made something read-only, there's generally no sane way back. At least if something's mutable, then you can still choose to not-change it. I believe that immutability, if necessary at all, should come with an option to either disable it, or to replace the object or value with a copy that can be modified. Consider what happens when I download an ODT; it'll open in OOo Writer, which will treat it read-only. This makes sense: from a user perspective, the document is on the web server, on which I don't have write access. It would only be confusing if I could change it and save that, as I would actually be saving to a temporary file and my work would be lost soon. However, I can save the document in my home directory and then edit it and save the modified copy again. A similar behaviour would be very welcome in some APIs. I feel the same way about constants too, by the way. Yes, it's nice if you can indicate that you intended the value to stay the same forever, but more than once I've used Perl's immense flexibility to override this decision, to bolt on extra functionality that the original author probably never thought of when in their infinite wisdom they decided that the value be constant. In other programming languages, or with complexer object oriented frameworks, one may not be so lucky and may not be able to say "(I think) I know what I'm doing, so get out of my way." Immutable (read-only) data is annoying in many ways, in many libraries and frameworks. The main problem is that once you’ve made something read-only, there’s generally no sane way back. At least if something’s mutable, then you can still choose to not-change it.

I believe that immutability, if necessary at all, should come with an option to either disable it, or to replace the object or value with a copy that can be modified.

Consider what happens when I download an ODT; it’ll open in OOo Writer, which will treat it read-only. This makes sense: from a user perspective, the document is on the web server, on which I don’t have write access. It would only be confusing if I could change it and save that, as I would actually be saving to a temporary file and my work would be lost soon. However, I can save the document in my home directory and then edit it and save the modified copy again. A similar behaviour would be very welcome in some APIs.

I feel the same way about constants too, by the way. Yes, it’s nice if you can indicate that you intended the value to stay the same forever, but more than once I’ve used Perl’s immense flexibility to override this decision, to bolt on extra functionality that the original author probably never thought of when in their infinite wisdom they decided that the value be constant. In other programming languages, or with complexer object oriented frameworks, one may not be so lucky and may not be able to say “(I think) I know what I’m doing, so get out of my way.”

]]>