Google Maps, now with billboards

Tuesday, October 25, 2011 

This is the first I’ve noticed location-specific ads being inserted into a google search. It makes sense that their map becomes a revenue source. If the ads don’t work, the the map will go away. It’ll be interesting to see how this plays out.

google location ads.JPG
ah… this
Posted at 15:05:25 GMT-0700

Category: technology


Tuesday, September 13, 2011 

I was searching for something random on Google (no, not that, regular expression examples) and noticed that funny little bar they put up there a while back when Google+ had the world all a-flutter. My little box had a [1] in it. Hmmm.. A few people I’d never heard of had “circled” me. Nobody I knew. I think I last checked G+ a few weeks ago, maybe it was a month. Oh well, so much for that one. Facebook will eventually do a MySpace, taking everyone’s cleverly crafted content out with it, but G+ won’t be the Facebook that does it. Or something like that.

Typing of Google, anyone else notice that Google has become much more aggressive about implicit substitution? I’m used to it autocorrecting typing, which actually led to ever more lazy typing, at least on my part. But I thought it always let me know when it was making a presumptive change. Search for [Congres] (using square brackets to denote the text box since “” has meaning in this context) and it used to note “Do you mean “Congress”” Yeah yeah, just fat-fingered the last letter, NP. Now it just silently corrects unless you use quotes. Maybe you actually wanted to find the “Hotel Du Congres.”

OK, annoying, but not fatal. But what is actually quite tedious is when you search for something slightly esoteric like [“white screen of death” client certificate]. 122,000 results. Whee. Oh, wait, most have nothing to do with client certificates – how can that be? [“white screen of death” “client” “certificate”] yields 367 results, almost all relevant. So for about 121,000 results Google assumes I just accidentally typed “client” and/or “certificate”? Those do not seem like common typos for [ ] (blank). If I went to all the trouble of typing out the words “client” and “certificate” does it not generally undermine the utility of a search function if it arbitrarily decides to ignore any inconvenient terms?

I find my self quote-forcing ([“white screen of death” +client +certificate] yields the same 367 results) most of my searches. Since when did my search terms become optional? WTF Google? Search is the one thing you do well. Well, that and advertising. Please don’t break it. Trust me, if you blow search you are not going to make up the difference with Social Networking.

Update: I recently searched for a scholarly article to back an assumption that document collections stored in structured databases can be accessed faster than document collections stored in file systems.  I used the word “median” rather than “average” in my search, but clever Google knows the two are often synonyms and rather than limit my search to documents that use the typically academic “median,” I got almost entirely useless results referencing various colloquial “average” constructions.

Posted at 01:37:41 GMT-0700

Category: Negativereviewstechnology


Sunday, August 7, 2011 

I’m sure it tastes fine, but don’t Google the name.
Posted at 00:10:34 GMT-0700

Category: funnyphoto

Will G+ Eat RSS, or Insist on Sole Ownership?

Thursday, July 21, 2011 

Weird: I have yet to find a way to import an RSS feed into G+. This is one of those things that significantly undermines Google’s “your data” cred. Anyone know of a way to do it? I haven’t found an “import RSS feed into your feed” the way facebook kinda does and the wordpress/facebook plugin does.

I’m a very strong believer in “he who owns the hardware, owns the data,” so, for example, posting this on G+ means that this text is Google’s (note, this was originally published on G+, then I stole it back!). And since it didn’t originate on my personal wordpress installation (free as in speech, free as in beer) running on my server at home (free as in speech, not absurdly expensive as in cheap beer), it isn’t mine.

My server also runs my mail server, my file server, my web server etc. all from my garage meaning that’s my data and my hardware and fully protected by law, while any data on Google’s server is effectively shared with every good and bad government in the world and my only legal recourse if it gets hacked or stolen or sold or given away or simply deleted is to… write an angry post on my blog and swear never to trust a cloud service again.

This is, obviously, exactly the same at FaceBook and every other cloud service. I use Facebook as a syndication service: I post on my own servers and syndicate via RSS to FaceBook, which becomes, in effect, the most frequently used RSS reader should people who haven’t gotten around to blocking me in their streams might find and by which perhaps occasionally be amused. This means I still own my data and my data has no particular dependence on FaceBook’s survival.

This post is visible only as long as Google wants it to be.  If Google changes the rules, I lose the data.  OK, I can download it – as long as they choose to let me, but it isn’t my data. When I post on my server then give FaceBook permission to republish the data, I control my data and they get only what I decide to give them. When I post this on Google and then ask “please, sir, may I recover my post for another use?” the power relationship is reversed: Google owns and controls everything and my rights and usage are only what they deign to offer me.

That almost everyone trusts the billionaire playboys who put king sized beds in their 767 party plane as “do no evilparagons of virtue is odd to me, but nothing better validates Erich Fromm’s thesis than the pseudo-religious idolatry of Google and Apple.  Still, even the True Believers should realize that the founders of these Great Empires are not truly immortal and that even if Google is doing no evil now, it will change hands and those that inherit every search you’ve ever done, every web page you’ve ever visited, every email you’ve ever sent, every phone call you’ve ever made or received, the audio of every message ever left for you, the GPS traces of every step you’ve ever taken, every text and chat and tweet might think, say, that Doing Good means something different than you think it does.  One should also remember the Socratic Paradox that renders tautological Google’s vaunted motto.

Unfortunately, at least so far, Google won’t let me use G+ to syndicate my data – they insist on owning it and dictating the terms by which I can access it. If I want to syndicate content through my G+ network, it seems I have to fully gift Google that content. I’m hoping there’s a tool to populate my “posts” from RSS so the canonical will remain on my server. Because it is the Right Thing To Do.

(Shhhh..  I’m going to copy and paste this into my own wordpress installation, even though I wrote it here on the G+ interface.  They probably won’t send me a DMCA takedown, but I do run the risk that they’ll hit me with a “duplicate content penalty” and set my page rank to 0 thus ensuring nobody ever finds my site again.  Ah, absolute power, so reassuring to remember that it is absolutely incorruptible.)

Posted at 11:47:35 GMT-0700

Category: politicsself-publishingtechnology

FB vs. G+

Tuesday, July 12, 2011 

An interesting artifact of the FB vs. G+ debate is the justification by a lot of tech-savvy people in moving to G+ from FB because they believe Google to be less evil.  It is an odd comparison to make, both companies are in essentially the same business: putting out honey pots of desirable web properties, attracting users, harvesting them, and selling their data.

Distinguishing between grades of evil in companies that harvest and sell user data seems a little arbitrary.  I’d think it would make more sense to use each resource for what it does well rather than arbitrarily announce that you’re one or the other.

However, if one is making the choice as to what service to call home on the basis of least “evil” and assuming that metric is derived in some way from the degree to which the company in question harvests your data and sells it, then it is somewhat illuminating to look at real numbers.  One can assume that the more deeply one probes each user captured by the honey pot, the more data extracted, the more aggressively sold, the more money one makes. The company that makes the most money per user is probing the deepest and selling the hardest.

From Technology Review May/June 2011, annual revenue per monthly unique US visitor:

Facebook: $ 12.10
Google:     $163.60

Google squeezes out and sells more than 13.5x the data per user. Google wins. But Facebook is gathering $12.10 worth of user data, why should Google allow Facebook to have it? If Google wins that last morsel of data to take to market and takes out Facebook, Google can increase their gross revenue by 7%.

I’ve also heard people argue that Zuckerberg seems more personally avaricious, mean, or evil than Google’s founders, comparing Google’s marketing spin to “The Social Network”

Zuckerberg’s only newsworthy purchase was a $7m house in Palo Alto. Google co-founders were in the news over a lawsuit between them over whether their 767 “party plane” (Eric Schmidt) could house Brin’s California king bed. This is in addition to their 757 and two Gulfstream Vs they talked NASA into letting them park at Moffet under the pretense that the planes would be retrofit with instruments for NASA. When they couldn’t do that (FAA regs, who knew?), they bought a Dornier Alpha, but still get to park their jumbo jets and gulfstreams inside NASA hangers for some reason. Suck on that, Ellison!

Posted at 01:25:13 GMT-0700

Category: technologyvanity sites

Moar Privacy

Thursday, December 9, 2010 

I’m using an Ubuntu VM for private browsing, and like many people, I’m stuck using a mainstream OS for much of my work (Win7) due to software availability constraints. But some software works much better in a linux environment and Ubuntu is as pretty as OSX, free, and installs easily on generic x86 hardware.

It is also pretty straightforward to install an isolated and secure browsing instance using VirtualBox. It takes about 20G of hard disk and will use up at least 512K (better 1G) of your system RAM. If you want to run this sort of config, your laptop should have more than enough disk space and RAM to support the extra load without bogging, but it is a very solid solution.

Installing Ubuntu is easy – even easier with an application like VirtualBox – just install virtualbox, download the latest ubuntu ISO, and install from there. If you’re on bare metal, the easiest thing to do is burn a CD and install off that.

Ubuntu desktop comes with Firefox in the tool bar. Customizing for private browsing is a bit more involved.

My first steps are to install:

NoScript is an easy win. It is a bit of a pain to set up at first, but soon you add exceptions for all your favorite sites and while that isn’t great security practice, it is essential for sane browsing. NoScript is particularly helpful when browsing the wacky parts of the net and not getting exotic browsing diseases: it is your default dental dam. Be careful of allowing domains you don’t recognize – Google them first and make sure you understand why they need to run a script on your computer and that it is safe. A lot of sites use partners for things like video feeds, so if some function seems broken, you probably need to allow that particular domain. On the other hand, most of the off-site scripts are tracking or stats and you really don’t need to play along with them.

BetterPrivacy is a new one for me. I am very impressed that it found approximately 1.3 zillion (OK 266) different company flash cookies AFTER I had installed TACO and noscript etc. You bastards. I’m sure I can enjoy hulu without making my play history shared-available to every flash site I might visit. Always Sunny in Philadelphia marks me as a miscreant. I flush the flash cookies on starting silently (preferences).

TACO is a bit intrusive, but it seems to work to selectively block tracking and advertising cookies. At least the pop up is comforting. For private browsing, I’d set it to reject all classes of tracking cookies (change the preferences from default).

User Agent Switcher is useful when you’re deviating from the mainstream. Running Ubuntu pretty much flags you as a trouble maker or at least a dissident. Firefox maybe a bit less so, but you are indicating to advertisers that you don’t respect the expertise of those people far smarter than you who pre-installed IE (or Safari) to make your life easier. Set your user agent to IE 8 because the nail that sticks up gets pounded down.

Torbutton needs Tor to work. Tor provides really good privacy, but is a bit involved. The Tor Button Plugin for firefox makes it seem easier than it really is: you install it and click “use tor” and it looks like it is working but the first site you visit you get an proxy error because Tor isn’t actually running (DOH!).

To get Tor to work, you will have to open a terminal and do some command line fu before it will actually let you browse. Tor is also easier to install on Ubuntu than on Windows (at least for me, but as my browser history indicates I’m a bit of a miscreant dissident, so your mileage may vary).

Starting with these fine instructions.

sudu gedit /etc/apt/sources.list
deb lucid main
deb-src lucid main

Then run
gpg --keyserver --recv 886DDD89
gpg --export A3C4F0F979CAA22CDBA8F512EE8CBC9E886DDD89 | sudo apt-key add -
sudo apt-get update
sudo apt-get upgrade
sudo apt-get update
sudo apt-get install tor tor-geoipdb

Install vidalia with the graphical ubuntu software center or with
sudo apt-get install vidalia

Tor expects Polipo. And vidalia makes launching and checking on Tor easier, so remove the startup scripts. (If Tor is running and you try to start it from vidalia, you get an uninformative error, vidalia has a “launch at startup” option, so let it run things.) Vidalia appears under the Applications->Network.

sudo update-rc.d -f tor remove

Polipo was installed with Tor, so configure it:
sudo gedit /etc/polipo/config

Clear the file (ctrl-a, delete)
paste in the contents of this file:

UPDATE: paste in the contents of this file:

(if the link above fails, search for “polipo.conf” to find the latest version)

I added the binary for polipo in Vidalia’s control panel, but that may be redundant (it lives in /usr/bin/polipo).

I had to reboot to get everything started.

And for private chats, consider OTR!

Posted at 17:45:45 GMT-0700

Category: politicstechnology

Testing Privacy Tools

Saturday, December 4, 2010 

I was curious after posting some hints about how to protect your privacy to see how they worked.

Using EFF’s convenient panopticlick browser fingerprinting site. Panopticlick doesn’t use all the tricks available, such as measuring the time delta between your machine and a reference time, but it does a pretty good job. Most of my machines test as “completely unique,” which I find complementary but isn’t really all that good for not being tracked.

Personally I’m not too wound up about targeted marketing style uses of information. If I’m going to see ads I’d rather they be closer to my interests than not. But there are bad actors using the same information for more nefarious purposes and I’d rather see mistargeted ads than give the wrong person useful information.

Panopticlik noscript.jpg

Testing Panopticlick with scripts blocked (note TACO doesn’t help with browser fingerprinting, just cookie control) I cut my fingerprint to 12.32 bits from 20.29 bits, the additional data comes from fonts and plugins.

Note that EFF reports that 1:4.1 browsers have javascript disabled. Visitors to EFF are, I would assume, more likely to disable javascript than teh norm on teh interwebz, but that implies that javascript-based analytics packages like Google analytics miss about 25% of visitors.


It is also interesting to note that fingerprint scanners (fingerprints as on the ends of fingers) have false reject rates of about 0.5% and false acceptance rates of about 0.001%. Obviously they’re tuned that way to be 50x more likely to reject a legitimate user than to accept the wrong person and the algorithms are intrinsically fallible in both directions, so this is a necessary trade-off. Actual entropy measures in fingerprints are the subject of much debate. An estimate based on Pankanti‘s analysis computes a 5.5×10^59 chance of a collision or 193 bits of entropy but manufacturer published false acceptance rates of 0.001% are equivalent to 16.6 bits, less accurate than browser fingerprinting.

Posted at 06:44:41 GMT-0700

Category: politicstechnology

Opting Out for Privacy

Friday, December 3, 2010 

There’s a great story at the wall street journal describing some of the techniques that are being used to track people on line that I found informative (as are the other articles listed in the series in the box below).  EFF is doing some good work on this; your browser configuration probably uniquely identifies you and thus every site you’ve ever visited (via data exchanges).  Unique information about you is worth about $0.00_1.  Collecting a few hundred million 1/10ths of a cent starts to add up and may end up raising your insurance premiums.

One of the more entertaining/disturbing tricks is to use “click jacking” to remotely enable a person’s webcam or microphone.  Is your computer or network running slowly? Maybe it is the video you’re inadvertently streaming back (and maybe you just have way too many tabs open…)

A few things you can do to improve your privacy include:

  • Opt out of Rapleaf. Rapleaf collects user information about you and ties it to your email address.  You have to opt out with each email address individually, which almost certainly confirms to them that all your email addresses belong to the same person.  You might want to use unique Tor sessions for each opt out if you don’t want them to get more information than they already have via the process.
  • Opt out at NAI. This is a one stop shop for the basic cookie tracking companies that are attempting to be semi-compliant with privacy requests.  If you enable javascript for the site (which would be disabled by default if you’re using scriptblocker) then you can opt out of all of them at once.  Presumably you have to return and opt out again every time a new company comes along.
  • Use Tor for anything sensitive.  If you care about privacy, learn about Tor.  It does slow browsing so you have to be very committed to use it for everything.  But the browser plug in makes it pretty easy to turn it on for easy browsing.
  • Don’t use IE for anything personal or important.
  • Run SpyBot Search and Destory regularly.  Spybot helps block BHOs and toolbars that seem to proliferate automagically and helps remove tracking cookies.  You’ll be amazed at how many are installed on your system.  I have used or not used TeaTimer.  I’m less excited about having a lot of background tools, even helpful ones than I used to be.  Spybot currently starts out looking for 1,359,854 different known spywares.  Yikes.
  • Check what people know about you:  Google will tell you, so will Yahoo.  Spooky.
  • Use firefox.  If for no other reason than the following plugins (personally, it is my favorite, but I know people who favor chrome or even rockmelt, but talk about tracking!)  Just don’t use IE.
  • Use the private browsing mode in your browser (CTRL-SHIFT-P in FireFox).  It’d be nice if you could enable non-private browsing on a whitelist basis for sites you either trust or have to trust.  We’ll get there eventually…
  • TACO should help block flash cookies.
  • Install noscript to block scripts by default.  You can add all your favorite sites as you go so things work.  It is a pain in the ass for a while, but security requires vigilance.
  • Install adblock plus.  It helps keep the cookies away.    It also reduces ad annoyance.  You can enable ads for your favorite sites so they can pay their colo fees.
  • Add HTTPS Everywhere from EFF. The more your connections to sites are encrypted, the less your ISP (and others) can see about what you’re doing while you’re there.  Your ISP still knows every site you visit, and probably sells that information, but if your sessions are encrypted they don’t see the actual text you type.  It also makes it harder for script kiddies to grab your passwords at the cafe.
Posted at 02:44:43 GMT-0700

Category: politicstechnology

the Cloud

Tuesday, April 27, 2010 

On the Media is an excellent resource always, but the second segment of the Apr. 23, 2010 goes over the lack of protection afforded data in the cloud due to the Stored Communications Act, an increasingly important topic.

Current law allows a very low standard for access to “Stored Communication” such as Gmail or Google Docs or any other “cloud service.” It turns out that Google gets about 20 requests for data a day and if an investigator asks for your email they do not need a warrant to get it.

If you don’t own the hardware, you don’t own the data.

Even if the Stored Communications Act is overturned, any data you store on a remote server such as Google’s, is Google’s and not yours. You have no right to get it back, no rights controlling Google’s dissemination of your data or resale thereof. In many cases there is a click through agreement with the service provider which may, for example, state that certain information will be kept private or not sold, but such clauses are typically superseded by statements claiming the right to rewrite the agreement without notification.

For example, FaceBook might change default privacy settings such that information you stored on their server with the understanding that it would be kept private is later exposed to search engines and indexed and thus made public, thereby increasing search traffic to their site, and thus to their advertisers.

FaceBook did not give, and was not required to give any particular notice. The data you put on their servers is theirs, not yours.

Don’t put data in the “cloud” you don’t want to be public. Google Docs is not a replacement for Open Office on your own hardware. Companies don’t make any money offering you free, private compute resources and storage; these services are profitable by exploiting the value of your information. In the long run it is probably cheaper to buy your own hardware.

Side note: in this excellent episode of OTM, they also cover the GAO’s pooping all over the MPAA/RIAA linkage between guerrilla antitrust (unauthorized copying) and economic problems. OTM also points out the linkage between the asinine ruling against the FCC and Net Neutrality, which is a free speech disaster, and worse still the MPAA/RIAA efforts to create a world-wide three-strikes rule to extort money to replace the money they used to be able to generate with their obsolete business model.

Posted at 20:55:03 GMT-0700

Category: politicstechnology