#musicbrainz-devel

How does the iPhone have nicer hardware? The N1 has a 1GHz processor (compared to 533MHz I believe) and a 840x480 screen.

the speakers on the N1 suck.

and the keyboard and typo correction on the iphone is better.

and the screen in direct sunlight is a bit a pain.

the automatic brightness adjustment on the iphone was much better.

check it: http://mayhem-chaos.net/blog

A lot of that is software though, not hardware.

I did a review after using it a week.

Hm

the light sensor in the phone SUCKS. and the speakers suck. those will never be overcome in software.

Eh, can't really speak from experience on the light sensor.

But I <3 my droid speakers

Now you got me curious. I'm gonna try to see if I can make a python script to mess with the light sensor.

I think the worst thing about my Hero is people constantly asking why my phone has a chin

hahaha

Tell them that it is so your phone can eventually grow a beard.

ha, if only

Beard growing is a major feature in Android 3.0, codename: zz top

if I wasn't ditching this phone soon I'd be happy just to get 2.0.1

Why are you ditching your hero? It is not a /that/ bad.

it's good but I'm after an N1 when they launch here

Ok, plane boarding time, I'm out. See y'all on the flipside!

http://xpc.xabbu.net/ Compare how the links work in the three sections at the bottom. Thoughts?

I don't see any difference between the left and the middle

The linkable area is different between the two.

oh, right

In the left one the entire div is a link, it's actually an <a> not a <div>.

I like the one on the right the most

Really? That's the one I like the least :(

I'm not a fan of the orange underline on hover.

I don't like how the left/middle ones don't act like a normal link (hover and nothing changes), and the left one... hovering over what looks like the page background and having it be a link is just messing with my head :P

hmm that one was probably the wrong one to demo that on. It's got the shortest text.

I'm not particularly attached to the underline though

What if I make the text and the image area all linkable? That would also include some whitespace underneath the text.

But it would leave out the h2 area.

i'll second you on that about the orange hovers lol.

I'm not really convinced, I prefer just linking the text :/

I removed the underline from the third box. I think it looks a bit better, but it's not very consistent with the rest of the links on the site now.

having no hover stuff at all like the middle one isn't consistent with the rest of the site either ;)

heh

But it's /radically/ inconsistent, so it's allowed :p The third box is just slightly inconsistent.

yeah yeah :P

Unless you have any suggestions for changing/adding content...

not at the moment

That sounds very um, interesting :)

pfft.

:)

time-of-day!

I should really stop being lazy and do some coding this weekend..

luks: yes! code things.

see warp code

code warp code

sea robot float

What if I can't see warp code?

haha

eek noisy.

mmmm, a 215MB install of Win95. Those were the days.

Although I'm sure you could get it in under 80MB or so.

djce: I think my first win95 installs were around 50mb.

sounds about right.

IIRC I have .tar.gz of a vmware disk of a win95 install at about 65MB

haha

I realised I had about 12 loose hdds lying around, with no idea which ones worked, which ones had any data on, etc. So finally having a clear out.

nikki: so, you would like to see the differences between libunac and the code we have in ngs, right?

yep

ok

nikki: http://dl.dropbox.com/u/5215054/unaccent-diff.txt

some of them seem wrong

I'll have to check if it's a bug in my code or just a difference due to newer unicode ve3rsion

except for some exceptions, the code *should* be the same as in libunac

which ones are you thinking of which seem wrong?

and thanks!

U+FACF (𢡊)U+FACF (𢡊)U+2284A (𢡊)

for example

or that ㉾ expands to two characters

http://dl.dropbox.com/u/5215054/unac-cmp.py is the code, btw

the first one looks ok, there's a bunch of characters which are encoded twice thanks to some crazy korean encoding which encoded each pronunciation separately even if they looked identical, and there's always duplicates simply because making sure the character you want isn't there before adding it is difficult when you've got thousands of characters

not sure what the second one is, I get an undefined glyph

oh right, a korean syllable... yeah, that seems fine too

haha... I just tried to do "perl unac-cmp.py"... funnily enough it didn't work

:)

all the characters I can remember wondering about are treated how I'd expect though, that's good

I wonder if that means i to ı and vice versa will finally be an auto-edit

hrm