wow ... won't fix. portrait mode is getting harder and harder to use. between 16:9 aspect ratios (1080 wide) and this "design decision" to *fix* the panel to the left side, you've probably only got 1024 useable pixels - cutting edge when XGA came out back in 1991

guess i'll try it on my laptop and see what i think. but on the desktop, it sounds brutal for anyone that prefers portrait mode - you get almost twice as many lines of code on the screen in portrait as you do in landscape, but those horizontal pixels become precious. i splurged and went 1920x1200, so i'm not in terrible shape - though the viewing angle is so bad, i have to keep the lcd turned 5 degrees

GNOME / Linux

not quite what you're looking for, but in the same vein. i'm using the nexus s on t-mobile prepaid with a google voice number

at my desktop: gmail voice chat over a 1Mbps, $10/month dsl connection
mobile with wifi: sip on the nexus s
mobile without wifi: t-mobile prepaid

everybody sees the same (gv) number. for sip i use callcentric + ipkall incoming, and anveo ($0.012 per minute) for outgoing. i'm using csipsimple instead of the builtin sip stack. i haven't tried sip over 3g. call quality with sip has been inconsistent, but i think i can improve it - i still need to tweak the echo cancellation params and figure out how to enable QOS on my router. i should also try skype-out

not a slam dunk, but calls are very cheap or free, sms is free - i spent $3 in the last month. when i need data it's $1.50 per day. it works because i'm at a computer most of the day, have wifi available most places i go, and don't make a ton of calls to begin with. at the price, it's hard to beat

this is a step in the right direction, but on top of that you need to layer something that allows persistence to messages, since the ad hoc networks aren't going to be able to route most of the messages that you want to send. maybe bittorrent or even smtp

A and B are in separate nets. A sends a message to B. everyone in A's net saves the message and when they connect to a new net they pass it on (to everyone in the new net). eventually someone finds B and delivers the message. as formulated highly inefficient, but should be tweakable to allow for short messages

been using google since beta, have 7 google accounts (that i'm working to reduce down to 2), have my own google apps domain, google voice is my primary number, chrome is my browser, perform dozens of searches a day

and had a workaround for an annoying problem ... google's use of synonyms makes it hard to search for something specific, appending "&nfpr=1" to a query disables it. even have a keyword search set up to automatically append it

and after all this time i learn that the "+" operator does exactly what i want without the kludge - thanks whitehaint - and i guess you can add me to the list of people that don't *really* know how to use google

from google.com/phone: nexus-s
Quad-band GSM: 850, 900, 1800, 1900
Tri-band HSPA: 900, 2100, 1700 HSPA type: HSDPA (7.2Mbps) HSUPA (5.76Mbps)

fcc filings seem to confirm (umts iv is 1700/2100) everything but the hspa-900

so the 850 is for gsm, not 3g
believe at&t 3g is 850 and 1900, so it looks like no 3g with the nexus s on at&t
the t-mobile version of the vibrant (on which the nexus s appears to be based) is umts 1700/2100 + 1900, and it's reported to get at&t 3g on the 1900
think umts 900 is for europe, but not sure how widely it's deployed atm

when the kernel accesses the slow disk, it is aggressive in trying to cache the read. if there's free memory this is obviously the correct thing to do, since if the memory is needed the cache can be dropped. but if memory is full, the kernel needs to decide whether to drop some file cache, or swap out a process. the default settings tend to favor disk cache, meaning every time you try to access anything on the desktop, the application has been swapped out and it has to wait for disk access to swap back in (often several seconds on my machine)

setting /proc/sys/vm/swappiness to a low value, eg 0, tells the kernel to favor processes at the expense of caching disk reads. this helps a lot in keeping the desktop snappy. kernel trap has a good summary of the issue and the developers motivations

swappiness doesn't help with applications that want to access a file repeatedly, but rely on the disk cache instead of an internal cache. eg, an IDE might have 10 source files in tabs, but instead of keeping the files in memory, it could just reread them each time a tab is switched. as long as the file remains in cache, this works fine. but when you copy a huge file, the source file gets dropped from cache, and the tab takes forever to refresh

not sure if there's an easy way for the kernel to know the difference between an application just copying a file, and actually reading it. but if there is, it would make sense to favor reads

if you're going to use your drive for a desktop or laptop, i'm guessing that your conclusion is correct. but for doing many small writes, eg a database, i think the "myth" is probably a very real problem. i haven't bought an SSD to test the real life behavior yet, but here's my back of the envelope calculation:

128G MLC drive, with .5M blocks --> 256k blocks. assume endurance is 10k. so after 2.5G erases the disk is at it's endurance limit. got to erase the block every time you modify it. 32M buffer can buffer the writes - assume writes are evenly spread over the blocks, and ignore write locations, so 32M buffer / 256k blocks --> on average 128 bytes per erase. 2.5G * 128 --> 320G of data. and the first write is free (since the MLC is blank). so 448G can be written before the drive is at it's endurance limit. 448G at 70M/s --> 6400 seconds of write at full speed, about 2 hours

so at least for a write-heavy databases, i can't see MLC being practical. might be able to work around this with a different architecture (and i'm toying with doing so in one of the projects that i'm working on). but for the traditional model, MLC looks like a non-starter, even for relatively small datasets

i'd love to be wrong on this, so correct me if i'm wrong ...

not sure about the distribution that the EEE used, but fiddled with my roommates ubuntu-based Dell netbook for a few days ... the ubuntu install was terrible. been using ubuntu for years. fuck, my parents have been using ubuntu for years. but with matt's netbook, i couldn't figure out how to do anything ... no idea how a windows user would have felt with it, but as a linux user, it sucked

agree with you to a degree, but it was pretty clear to me where he stood on IP and such when he selected biden as running mate. as much as i don't like where he's headed, i don't think we can claim that he's changed directions much

actually, i'd say he's been remarkably honest to the message that his campaign sent (both explicit and implicit) ... just too bad that i didn't agree with too much of that message :(

the problem that i see is that there's no accountability. same thing for the tea party. or for that matter, obama. i gave a bit of money to the campaign, and there might be a lot of people that did that don't agree with some of his decisions. but it's very hard to apply any sort of leverage - the campaign is the single point of contact, and as such largely controls the message

for the sake of argument, lets assume that 30% of the money that the obama campaign raised was from people who actively oppose the speech the article is referring to. (yes, i know it's probably _far_ lower ... this is for the sake of argument) it's almost impossible for that 30% to find each other, to figure out that other "supporters" aren't happy with the current direction. if that block could get together, they'd have some influence. but we can't

i supported the obama campaign largely because the financing was (at least initially) driven by small donations. that concept has the potential to be revolutionary, but i think there needs to be another step. the people making the donations need a way to maintain some control, to have a unified voice

we need a proxy - to not give money to a campaign directly, but thru a proxy that represents a particular set of beliefs - and maybe even to negotiate with various candidates to try to find the best match. the coffee (or tea) party might be a step in that direction, but it seems to be more about positive energy than any particular issue (or at least, not issues that i'm strongly in favor of)

there - fixed that for you. google chrome runs great on ubuntu 9.04 - quick startup, fast page rendering, fast switching between tabs, and flash works fine. i've had a page crash, but never lost the full browser - usually have 5-20 tabs open. chrome's "developer tools" are good - firebug is better for many things, but the developer tools are much less intrusive, don't slow down page loads nearly as much. the two tools compliment each other well

in short, chrome is a big improvement for me ...

i guess i'm the evil programmer type - i read the article for the sake of seeing some clever tricks. i stopped reading after his first example ... not clever at all, not obscure at all, no need to comment, no need for fancy variable names, no nothing ... and i don't think i've coded in c in 5 years

for(ss = s->ss; ss; ss = ss->ss);

if there's any ambiguity about what that line is doing, you're not qualified to be commenting on anyone else's code

not sure that usc belongs in that first list ...