Oh. I guess the reason no one’s left a comment on my blog in ages is that there isn’t a comments form.
I should fix that… today, probably. Right now I’m tired.
Oh. I guess the reason no one’s left a comment on my blog in ages is that there isn’t a comments form.
I should fix that… today, probably. Right now I’m tired.
Does anyone know if/how I can (in Linux, of course) control sound volume per application? So I can, e.g., turn down the sound coming from a flash game in Firefox, but let Rhythmbox play my music at normal volume?
So I was thinking about ad-blocking hosts files today, and it occured to me: how come ISPs don’t have ad-blocking DNS? It’d be a big selling point…
What happens if you serve an SVG document (SVG namespace, only SVG elements) as application/xhtml+xml? What should happen?
(“Happen” of course meaning semantically.)
Maybe once I’ve finished HTML5, I can work on a new SVG to go with it.
Oh please please please yes
I have one profile I use normally, one I use only for a few things (with special settings, userContent.css, etc.), one I use when I need to be logged in to the same site with two logins at once, and one I use for testing that I constantly delete and re-make to clean it.
To use them, I have a main button in the panel that just calls path_to/firefox, and choose the profile I need, or if I need two at once I open a terminal and type path_to/firefox -no-remote. This fails in a small way in one particular circumstance: If the partial update has already downloaded, opening a second instance will install it, which can lead to minor breakages (XML error in download manager, e.g.) in the first. Other than that, though, it works perfectly for me.
You know you’re a geek when you think “My room’s cold. Perfect time to compile OpenOffice.”.
Microsoft is trying to kill OpenGL. The Web is embracing it. Who will win?
You know what would be a really nice extension for bug testing in Firefox? Something that hooks into the profile manager and lets you create a one-use profile without having to go through the four-step “delete old clean testing profile/create new/type in name/finish” sequence. When you’re looking for a regression range and opening with a new profile each time over and over, the savings would really add up.
For the past few weeks I’ve had no sound coming from my computer. I switched over from having sound support built in to the kernel to loaded as modules, and everything worked fine. Then, just to be sure, I re-ran the configuration… and it broke. Every single step in the guide I was using worked fine, with no errors at all or anything, until the part where you were supposed to hear sound, which I didn’t.
Lack of an error message is a pain to work with, so I couldn’t find any way to fix it. I went through the guide several times, it didn’t get fixed. I upgraded kernels and alsa tools, nothing helped. I asked on forums, no one knew.
So today I tried everything again, and then went and searched the alsa-user mailing list archives. And there I found the solution to my problem: switch a rather buried, nowhere-mentioned pref to change the output from digital to analog, since it can only do one at a time.
So hopefully this post will come up on google for anyone having the same problem as me. If you have a Soundblaster 2 Audigy (ZS (Platinum (Pro))), which uses the emu10k1 driver, and you’re not getting any sound, but you have no error messages either, and you’re sure everything’s turned on and unmuted and turned to a good volume, then scroll way over in your alsamixer preferences, and find the thing labelled “Audigy Analog/Digital Output Jack”, and unmute it.
Short track speed skating is the most awesome winter olympic sport ever.
So I’m reading the comments of a fastback memory usage post on Ben Goodger’s blog, and people are saying they rarely use the back button more than one or two times in a row, and storing 8 pages is maybe a little excessive.
And I just realized, I often use it not once or twice or eight times, but twenty or more. You see, I have a few html-based games I play, and when I’m moving around a map I can be loading over a page a second. And when I want to, say, go back and see the results of the last battle, that can be quite a few pages back even though it happened less than a minute ago.
Contradict that to slashdot where I open each story to read in a new tab and never use back at all, or google where, if I think I know what page I’m looking for but I’m wrong, I use it exactly once at a time. All in all, I think it’s pretty obvious that with such a wide range of usage even from just one person!, some sort of pseudo-intelligence is needed here.
As a disclaimer, I’ve never noticed any memory problems at all, though. Perhaps that’s because I have a gig of RAM and use Linux, and Linux’s memory management and multitasking/multi-userness generally pwns Windows’.
I really must start posting on www-html more, instead of just reading. As far as I can tell, the HTML WG hasn’t released a good spec since HTML 4. (And that I only count “good” as it was an obvious improvement over HTML 3.2; everything since then has been stagnant or worse in terms of good design.
Also, I need to post to the WHATWG list more, too. While it’s making definite improvements with HTML 5, it still doesn’t separate documents from application UI. That makes it “relatively good” as HTML 4 was, and non-relatively probably the best spec of its type there ever has been, but still not good enough.
(This post prompted by something Anne said.)
Which is better: localized content, or localized presentation of content?
For example, which is better:
Serving
<span class='date'>February 2rd at 8:30 PM EST</span>
to me, and equivelant things to other users, or
Serving
<style>.date{binding:datelocalizer.xbl;}</style>
<span class='date'>2006-02-03T01:30Z</span>
to all users
What is the semantic difference between the following:
<html xmlns='http://www.w3.org/1999/xhtml'>
<head>
<title></title>
</head>
<body>
<p>This is a paragraph.</p>
<object>
<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 2 2'>
<circle cx='1' cy='1' r='1' fill='lime'/>
</svg>
</object>
</body>
</html>
<html xmlns='http://www.w3.org/1999/xhtml'>
<head>
<title></title>
</head>
<body>
<p>This is a paragraph.</p>
<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 2 2'>
<circle cx='1' cy='1' r='1' fill='lime'/>
</svg>
</body>
</html>
<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 2 2'>
<foreignObject>
<p xmlns='http://www.w3.org/1999/xhtml'>This is a paragraph.</p>
</foreignObject>
<circle cx='1' cy='1' r='1' fill='lime'/>
</svg>
<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 2 2'>
<p xmlns='http://www.w3.org/1999/xhtml'>This is a paragraph.</p>
<circle cx='1' cy='1' r='1' fill='lime'/>
</svg>
(Feel free to correct my SVG, but that’s not what the question is about.)
Weird. I just noticed that with the new frame painting stuff from Bug 317375 even though most scrolling has slowed down, Hixie’s blog, which was always slow, has sped up tremendously. How strange.
div, in current markup languages, has two uses:
The end.
Overuse of Floats Considered Harmful definitely deserves a link. I’ve thought the same thing for a while now, and do my best to avoid using them except for things like floating pictures in columns of text. You can tell, when you try to use them for overall page layout, that they just weren’t made for it and attempting it is hackish at best. Hopefully the new layout features of CSS 3 will arrive soon, and soon enough that float usage isn’t hardcoded into authors’ minds like table layout has been.
So why haven’t I heard of any CMSes using Bayesian filtering to automatically suggest tags for blog posts? Am I reading the wrong places? Have I heard of it, and just have a bad memory? ‘Cause this can’t be a new idea…
Bad Behavior has blocked 54 access attempts in the last 7 days.