Slow afternoon thoughts: 3 links

[da]

I seem to have a pile of things bouncing around in my head, mostly in the not-useful state. But this bit has shaken out, and I want to record it for my own sake, if not for anybody else's.

Remember that "Pi" song and video? The band has an entire album, and a few more fun videos on their site. (Warning: these links are currently worksafe, but much of their site isn't. Just sayin'.) The music is mostly cynical bluegrass, which sounds to me like an untapped market. The song about patriotic dinosaurs is great, and so is the "matrix" video set to melodica.

The second musician link: Jonathan Coulton is a guy with a guitar. Kind of a musical xkcd, if you will; a lot of his themes are wistful and pretty geeky. He seems to have a following, and it seems to be mutual: one guy turned his podcast series into a book and a colouring book. Another made a card-game based on his song "Code Monkeys." Anyway, I like his science stuff. And "Re: your brains" is fun; it's about zombies at the office.

Both these bands release music through cdbaby, which has a scary, scary huge assortment of artists I've never heard of, most with samples of their music. CDs mostly in the $10-15 range. They have neat categories for finding music. ("If you're depressed and want to stay depressed", etc.) Where to find the time to browse it all, though?

Finally: the google-borg continues taking over my internet experience. I've been using google's web history for the last month and it's just great. It's an opt-in to let google record the URLs of where you've browsed.

But google also already has the page contents, for things they index; so in effect you can do a google search that only matches pages you've already visited, which turns out to be something I want to do every few days or so. Of course it also includes prior google searches and search output pages. Quite useful.

It requires the Google Toolbar, which has a "no visible toolbar" option so Firefox looks the same, except for the search box, which is slighly google-ified. And with an unexpected feature that the search box "suggested searches" are also useful (eg., catch misspellings before I hit "return").

Since it works on mac/linux/windows, my work and home browsers are recorded somewhere I can get to from anywhere with the net.

Oh right; it also has a "suspend recording" option and you can delete records from the history. And it seems to not record pages that aren't already in the google index; so password-protected stuff is safe. This privacy/usefulness tradeoff is acceptable to me.

Comments

[chezmax]

Suggested searches work in my firefox without any google alterations. It'll even do google calc on the fly (type 3*5 and wait a moment, and suggestion box pops up = 15).

[da-lj.livejournal.com]

Ah- I'm guessing you're running a newer ff than I am. I'm sticking with 1.5, b/c I like it more than 2. :)

[kraig.livejournal.com]

No more security patches for 1.5, eh?

[da-lj.livejournal.com]

Rats. They just ended support in "mid-May". Oh well, so much for that run. ;)

[psychedelicbike.livejournal.com]

Wicked! I just typed "e" into the box, and I got "= 2.7182818..."

[dr-tectonic.livejournal.com]

Johnathon Coulton's cover of Baby Got Back is superfantastic.

And I'm SO going to look into that Google webhistory thing. That is EXACTLY what I need.

Jonathan Coulton

[bats22.livejournal.com]

I love his "Code Monkey" and "Baby Got Back" covers. FYI, the New York Times Magazine did a story that included some time spent with him:

Sex, Drugs and Updating Your Blog, Published: May 13, 2007

Re: Jonathan Coulton

[da-lj.livejournal.com]

Oh /awesome/.

Re: Jonathan Coulton

[da-lj.livejournal.com]

That's a great article. Great symbiosis model:

Coulton’s fans are also his promotion department, an army of thousands who proselytize for his work worldwide. More than 50 fans have created music videos using his music and posted them on YouTube; at a recent gig, half of the audience members I spoke to had originally come across his music via one of these fan-made videos. When he performs, he upends the traditional logic of touring. Normally, a new Brooklyn-based artist like him would trek around the Northeast in grim circles, visiting and revisiting cities like Boston and New York and Chicago in order to slowly build an audience — playing for 3 people the first time, then 10, then (if he got lucky) 50. But Coulton realized he could simply poll his existing online audience members, find out where they lived and stage a tactical strike on any town with more than 100 fans, the point at which he’d be likely to make $1,000 for a concert. It is a flash-mob approach to touring: he parachutes into out-of-the-way towns like Ardmore, Pa., where he recently played to a sold-out club of 140.

As they point out, the tricky thing is if you want to continue growing without a label. I wonder how likely it is that the labels will adapt to promote a more even partnership for mid-size self-grown bands (who still insist on giving away DRM-free music!) Interesting times.

[mynatt.livejournal.com]

Oh right; it also has a "suspend recording" option and you can delete records from the history. And it seems to not record pages that aren't already in the google index; so password-protected stuff is safe. This privacy/usefulness tradeoff is acceptable to me.

This still freaks me out. I know Google could already be tracking my searches anyways, but I'm just not ready to become one of the borg quite so fully yet. Then again, they say resistance is futile...

[da-lj.livejournal.com]

Heh. Or to quote secretsoflife: "We are the Canadian Borg. Resistance would be impolite. Please wait to be assimilated. Pour l'assimilation en francais..."

I guess I find my web searches too boring for google to really care about me in non-aggregate. Even if they paid attention to the bits that I told them to not record. It would be a completely different story if this was a remote cache of /everything/ including password-protected pages. That would be uncool.

The part that freaks me out about the googleborg would be a concerted effort by someone to guess my password... but I'm not sure what they could do with that data; these days, I'm not a terribly blackmailable person. :) The worst they might do is delete stuff. Or send email as me, which as I think about it, they already could spoof.

[mynatt.livejournal.com]

I'm mostly worried about the idea of large-scale profiling. I don't think I'm doing anything worth blackmailing me for either, but the idea of Google learning about me (even in aggregate) and predicting what I'd like to learn about, and giving me helpful suggestions, is scary. I don't want to outsource my curiosity to them! Also, if I lived in the US I'd worry that the FBI or NSA would secretly start datamining everyone's search habits, and the smaller their database is the less tempting a target it would be. I realize I'm probably just being paranoid here, but I really think it might happen one day.

[da-lj.livejournal.com]

I don't want to outsource my curiosity to them!

*nods*

...that leads me to an interesting image; the followup to asian kids levelling up WOW characters for cash: reading blogs and submitting recommendations to digg or something. Which I bet people are doing for cash, too.

On the non-governmental side: for whatever reason, I'm not troubled by google's aggregation, as I am (say) by Amazon's. Probably because google feels more like a middleman who stays out of the way and carefully delineates where things are "commercial". Amazon feels like someone following me around a store trying to make recommendations, which is damn creepy. ...Plus they rewrite pricetags when you're not looking.

Of course, google's "non evil" front-end includes a huge pile of potential evil uses as well; such as searches from China being "optimized" for Chinese legality.

I'm taking it on faith that google won't become increasingly more evil over time, because if they do, they've got the data for it.

On the governmental side, I expect the US government is already data-mining at main router exchanges without google's help. I expect I should be more worried than I am about that.

[da-lj.livejournal.com]

Methinks I spoke too soon.

http://www.ft.com/cms/s/c3e49548-088e-11dc-b11e-000b5df10621.html

Asked how Google might look in five years’ time, Mr Schmidt said: “We are very early in the total information we have within Google. The algorithms will get better and we will get better at personalisation.

“The goal is to enable Google users to be able to ask the question such as ‘What shall I do tomorrow?’ and ‘What job shall I take?’ ”

Oh oh.

OTOH, the article points out that google will only keep 2 years of user search information.

[mynatt.livejournal.com]

What really scared me was Google getting subpoenaed to hand over individuals' search histories. That was before "iGoogle", and they refused on the grounds that it would be too hard to extract one person's records from their logs. I doubt they'd be able to claim that now. Having all that information on so many people available for seizure, either as part of a lawsuit or nationalized en masse to fight The War On Terror(TM), is just too tempting a target for the state to never try to take it.

[da-lj.livejournal.com]

they refused on the grounds that it would be too hard to extract one person's records from their logs.

I'm not sure what I think about that on a individual person basis. I'm convinced that subpoenaing for 1m users at once is illegal.

I was just googling around (of course) to get more info on this sort of thing. I googled "google refused subpoena", and only came up with the big Jan '06 case:

http://www.boston.com/news/nation/articles/2006/01/21/google_subpoena_roils_the_web?mode=PF

...which seems to suggest that google's objections include the argument that complying /without/ releasing individually identifying information would be too difficult. And that fishing expeditions were illegal violations of privacy..


Google is vowing to resist efforts by the US Justice Department to obtain information about the searches run by millions of its users, even though investigators are seeking aggregate data about Internet use, not individual users' records. The Justice Department wants the information as part of its effort to defend the Child Online Protection Act, a 1998 federal law that seeks to ban Internet sites from displaying content that the government deems ''harmful to minors." The Supreme Court has ruled that the law can't be enforced unless the government shows less intrusive measures such as Internet filtering are inadequate. The government hopes to use search results from Google and other companies to show that Internet pornography is so pervasive that only a federal law can protect children from it.

Yahoo Inc., Microsoft Corp.'s MSN search service, and Time Warner Inc.'s AOL service have all agreed to provide the information, according to a Justice Department spokesman. But Google has refused, saying that releasing the data would compromise its users' privacy and the company's trade secrets. ''Google is not a party to this lawsuit and their demand for information overreaches," said Nicole Wong, Google's associate general counsel. ''We intend to resist their motion vigorously."
[...]
In papers filed yesterday at a federal court in San Jose, Calif., government attorneys said that they are not seeking information about individuals. They want the search companies to provide a sample of a million websites from the billions they currently index, as well as all the search terms typed into the services during a one-week period. All information that could identify individuals is to be removed before the data is given to the government. The government could use the data to estimate how pervasive pornography is on the Internet and how often pornographic sites come up in random Internet searches.


The interesting development there for me was that google was still fighting that subpoena, a month later, when the ACLU was saying they'd have to subpoena google for the same info if google caved, as they had AOL and MS, which fed into Google's arguments against submitting to the huge subpoena.

http://news.com.com/2100-1028_3-6041223.html

Interesting stuff. ...I wish aclu's website wasn't down right now, I'd try and track this to the end.

I'd like to think more about subpoenas and individual search histories... but I think I've settled that as less worrying to me than mass subpoenas. Which isn't what you were talking about, I realize. :)

[da-lj.livejournal.com]

...and to continue my diversion, it seems google avoided the subpoena 'till the report was published:

http://sethf.com/infothought/blog/archives/001095.html

V. SUMMARY

This study reports on the Google and MSN indexes, on AOL, MSN and Yahoo! queries, and on the most popular Wordtracker queries.

[mynatt.livejournal.com]

Ah, okay. I had misremembered the details of that case. I'm glad that Google fought it, and that they won. I still have more respect for them than their competitors, ethically as well as technically. For example, in China, Google is the only company that doesn't save any data on users, so for example they can't even be asked to turn over information which would imprison people (as Yahoo has done). They're also the only company that tells you when a search result has been censored, which is nice.

But I still worry that one day, they might not win in court, perhaps after some sort of special bill has been passed by Congress. I'm not so much worried about the consequences of my own data being turned over, as that I want to make their dataset smaller and thus less tempting to investigators. I fully realize this is largely a symbolic gesture, though. :(