Open Hatch

Archive for the ‘Techie Stuff’ Category

I was just fiddling with the search toolbar and searching from the address bar in Firefox. Somehow I ran into pages detailing how to alter the search behavior of the search bar in firefox so that it doesn’t automatically direct you to google’s “lucky” (#1 ranked) result of whatever you type. (I think that default behavior is presumptuous and annoying.) I also ran into a page with links that will add scroogle.org (for a variety of regions, and for SSL connections, too) to the search engines available from Firefox’s toolbar.

(For an explanation of scroogle.org, go to their web page.)

Using the latter, I figured out the format for queries through scroogle SSL. With this, a search can’t be tracked by google, and can’t be deciphered by anyone (your location can be found, however. Unless you add in anonymizing proxies; then the only trace would be remnants on your hard drive, unless you use Private Browsing mode in Firefox, and/or encrypt your whole drive with TrueCrypt).

Following then is a URL for entering into step 4 of said instructions, so that by default, searches from the Firefox address bar will query through scroogle. To do it without SSL, enter this URL in step four of the instructions at this page:

http://www.scroogle.org/cgi-bin/nbbw.cgi/search?q=

If you want SSL (so that the search itself is impossibly obfuscated to any eavesdropping intermediary), just ad an s to http (https), so that this is the URL to enter in step four:

https://ssl.scroogle.org/cgi-bin/nbbwssl.cgi/search?q=

The real practical application of this may be limited, but it pleases my inner nerd

Modifying address bar behavior aside, I do think adding different search engines to the search toolbar is very practical. There are many varieties of information (and scenarios for their use) where the findings of a ranking algorithm are not ideal. To that end, here is a page that will lead you to any (or many) links you choose, which will add useful search engines to your search toolbar (not only to Firefox, but to unrelated browsers as well, I think!). And here is an associated (very cool!) Firefox add-onn that will add the search capability of any web site to your search toolbar; so that you can skip navigating to that site’s search tool, and just search from the toolbar.

I’ve been trying to get CorelDraw X3 working on a PC to edit .svg files for a project (wow, the svg format is cool). For some reason, it’s arguing with this PC, although it works fine on another (much slower) of mine. Every time I go to the file open menu it brings up a crash report dialog box. I can cancel the dialog and the program continues running fine – after it appears 3 more (total 4) times.

I love this application, but what in the world is with this? I finally tried the obvious, duh approach to finding a solution: google it. I quickly found a page with a proposed fix (from a user with a serious chip on his shoulder).

..You must launch regedit (ALT+R or START->RUN and type regedit and press enter) and go to:

[HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Policies\comdlg32]

and change NoFileMru from 0×00000001 to 0×00000000

Registry hacking? Oh, brother. However, it worked. Problem solved. Application running normally again.

Since that person’s post saved me grief, I’m paying it forward. You can do the preceding registry hack much faster by using a .reg file. Download the following text (.txt) file, rename the extension to .reg, then double-click it. If Windows doesn’t report successfully importing the information into the registry, try right-clicking the file and then clicking “merge”. These instructions are for a Windows XP account with Administrator rights; Vista users may need to disable User Account Control or run it as an Administrator or whatever.

Registry Fix file (.txt to rename as .reg)

Creepy oddness I just ran across: after a friend of mine on Facebook shared a link, their friend saw an advertisement on CNN.com that said so. “[So-and-so] shared [this-and-that] on facebook”, it said. This is probably only targeted to friends of that friend. Presumably, this is a service, or else why would CNN and Facebook be in kahutz to set this up without either user’s awareness or express permission?

How (technically) did it happen?

By default, Facebook allows all of the following information about you to be shared by your friends:

• Personal info (activities, interests, etc.)
• Status updates
• Online presence
• Website
• Family and relationship status
• Relationship details (significant other, looking for, etc.)
• Education and work
• My videos
• My links
• My notes
• My photos
• Photos and videos I’m tagged in
• About me
• My birthday
• My hometown
• My religious and political views

“Shared” means publicly available for harvesting by facebook applications, or now, apparently, facebook partner web sites (read on). If you’re logged into facebook, here is the link to the friends privacy settings page where you can uncheck all that.

That isn’t the only place with related settings, it seems. This page has a setting for whether “select partner” web sites (which evidently includes CNN) can “personalize your experience” by use of such information.

Even if you disable that, and you have a facebook account and have “added” any reasonable amount of people you know, the following is true: anyone can find your profile page, and from that page, they can learn all the following: your name, gender, “networks”, friend list, and Pages. This (minus a few things you can withhold, like a profile picture and your city) is all defined by Facebook as “publicly available information“.

All of this bothers me. About their “publicly available information” policy, I agree with this blogger – this search of his blog pulls up three posts about it at this writing. The report he links to from the EFF is alarming, and frankly damning. And I never thought I’d agree with the ACLU on any point

It seems to be the most common attitude of social media engineers; that there is no ethical problem in manipulating people into disclosing as much about themselves as possible, and making this information boundlessly and permanently available. They declaim: we respect your privacy. Rubbish! Their bottom line is advertising, and that means the more information about users they can harvest and exploit (with or without your knowledge and express consent), the better their bottom line. End of story – and all the evidence in the ongoing story supports it.

Google Buzz is also terrible with privacy. Here’s the revision as of this writing, of the Wikipedia entry’s section on privacy concerns over that service. Most notable and alarming there is the report that by using Buzz, a woman was found by her abusive ex-husband, because it shared her contact and work information without her knowledge or consent.

This page relates how to disable the service in gmail, or at the least curb what it discloses about you.

Other readings – a blogger who was creeped out by Buzz’ initial release – Google quoted as officially stating “there is no complete privacy” – a blog post that may be fairer to Google. Still, the release of Buzz was recklessly dismissive of (or even contemptuous toward) privacy concerns.

I’m kinda puzzled, though. The newest relevant blog article I can find on it (and here are two previous – 1 – 2) relates that Buzz hasn’t disabled default sharing of your “following” and “followed by” lists, but; in my own test of Google Buzz, after I joined, activated my profile, and logged out, I viewed my public profile link, and it didn’t display who I’m following and who I am being followed by. (I’ve suddenly just imagined a strange and tasteless modern retelling of Jesus’ arrest and mock trial, in which Peter is accused by those around him of following Jesus’ twitter account, and denies it thrice before, erm.. the google cache reveals it.. and Peter goes out and weeps bitterly.) It also said there wasn’t enough information to be indexed, so maybe that’s why. Regardless.. it’s well enough for me to be part of Google’s Brave New World via the search engine and gmail. Sigh – and YouTube.

I’ve kept getting this error message at my Ussins blog whenever I try upgrading wordpress automatically, and just got it now when trying to automatically upgrade plugins:

“Fatal error: Allowed memory size of 33554432 bytes exhausted”

Fed up with seeing it, I googled it and was led to this page, with a simple fix recommended by the commenter “gestroud”:

You could also add this line to your wp-config.php

define(‘WP_MEMORY_LIMIT’, ‘64M’);

This way you won’t have to constantly make the fix again whenever you upgrade WordPress.

I added that line near the top of said file in the install, but made it 128M (for 128 Megabytes), and tried the plugin upgrade again. Error fixed.

Someone else there mentioned it might be caused by a php.ini in the wp-includes folder that would have been added manually “by you or your host”. I find no such file at my web server, only a php.ini.default, and I don’t know if that applies. If it did, it would be changing the line that starts:

memory_limit 32

Which for me had a byte listing (not 32 as in the comments in that thread); I’d assume it would just be quadrupling that number to equal 128 megabytes.

These children were born too long ago; it is time to release them.

I’ve created a set of Windows batch scripts that retrieve (over the internet) and cross-breed Electric Sheep genomes, which I’ll show results for here. There are two galleries – scroll further down for the second.  The batches retrieve the sheep genomes (or instructions for creating these images) either by checking against saved image names or at pure random via the random.org number service.

The first gallery is of the parent genomes, and the second is of their children, nearly 500 of them, which survive many thousands of aborted children. They are interbred by both hand-picked and randomly picked genomes.  Many of these are rendered at a very high resolution of 2560 x 1960! Altogether, rendering these took a Pentium III machine working non-stop for about a week.

It is my intent to add these into the mix of my mass-customization product picker; I need to work out credit and payment sharing with generation 242+ sourced images. Images that say .242 in them are not free to reuse; everything else is. Note the links to show any showcased image in full, huge-resolution size

Here are the parents:

And here are the children:

I ran across this site, foursquare. That link is to an anonymous (to me) user who volunteers exactly when and where he is, at any and many times. Specific times and addresses, and all of this is completely public. It is creepy enough knowing the founder of Facebook makes breathless declarations about how (allegedly) outmoded privacy is, and how easily people could be burglarized as a result of tweeting about their vacation (or updating their status in any of a variety of social media).

WARNING: very foul culture exposed in the first link to follow – avoid if you’d rather not know.

[link 1 - link 2 - link 3.]

The article at the first of that group of links explains how easily you can locate any stranger’s house by connecting, as an example, geotags from a flikr account to an individual. But this foursquare service provides all that in one glance, in addition to time information in one glance, for anyone to look up. Any burglar (or worse) who wants to find a target doesn’t even need their name. Time and location is enough.

mindlesstweeter plane landed! did u see my tweets about my new expensive electronics? pls burglarize me!

goods4me @mindlesstweeter: in ur house raiding ur expensive electronics! u cannot find me my accnt is behind anonymus proxy hahaha

I am downloading so many gigabytes of abstract art animations from where someone has uploaded their repository of Electric Sheep movie files (I’m in contact with this person; I’ll be uploading the ones I have for him – and for you! – to access). As I do this, the download eventually runs into an error: out of hard drive space.

Yoink!

I’ll need to move what I’ve downloaded to an external hard drive to free up space. So I connect the drive and start doing this. The computer hangs (several high octane applications open, music playing, and many high octane data transfers will do this – if your computer is a few years old). No usual attempts to unfreeze it succeed. Finally it occurs to me it’s the data transfer that is probably the real holdup; I’ve seen this setup unfreeze before if I simply disconnect the external drive to interrupt it. I disconnect it, and the instant I do so, everything else on my computer is freed up, including my music player, which proceeds to the next song in my queue, which it happens is not a song, but a video game sound effect, and this sound effect besides:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

How appropriate.

If you have not enjoyed this happenstance, you may take reprieve in the idea that it is possible you may not be a nerd. And/or that you have never played (or fully understood, as it becomes any human beings’ divine duty to understand) EARTHBOUND, the classic among classic Super Nintendo video games from which this sound effect comes, at a moment at which a very important something (someone), a bee, dies. Like.. like.. like a failed data transfer. Oh, the poetry.

At opentochoice, “choice matters”:

“..the Web browser has become one of the most critical and trusted relationships of our modern lives – with nearly perfect knowledge of everything we do.”

Um, no.

.. And I’m thankful for the Mozilla Foundation, and search engine optimization, and my search engine ranking, and Firefox plugins.. and please bless that Google will stop nagging me to opt-in to Google Wave..”

The ‘net is great (even arguably crucial), but this sounds like.. actual worship. Wrong god. Idol Fail.

(I actually am thankful for the Mozilla Foundation, though.)

I’ve been wanting something like this for some time and may have finally found it. This tool is great:

tag2find

After installation you can “tag” any file with a right-click, then left click “Quick Tag this..”. To “tag” a file means to associate one or more words or phrases with it. Then this tool lets you search for files you have tagged. If you copy or move the file, this background program automatically updates its database – it keeps track of the file for you. This is very useful if you move files around and otherwise have a hard time keeping track of them, or simply don’t like the dozen mouse clicks otherwise necessary to peruse your folder heirarchy

Better still, when you install it, it asks you where you want its tag database to be located. That way, when you inevitably must reformat the Windows hard drive and assuming you maintain the good (nay, crucial) practice of backing up your files to a separate hard drive, this database will still be available and usable.

Another implication of determining the database location is (blathering now) show

The advantage of using an external database is there is no fuss with the varying, incompatible, dysfunctional standards for tagging files themselves. The tag data is external to the file yet perfectly managed in reference to it. Even when you can tag a file in Windows, this often uses file system informational extensions which can get wiped out if you copy the file to another drive (or other media), or worse, if they are copied elsewhere, it may be when you don’t want them to be.

The program runs in the background, apparently does not produce any noticeable slowdown, has a minimalistic and great user interface, and it can integrate with the Windows shell right-click menu (which is what allows you to right-click a file to tag it).

My initial impressions of this tool are very positive – I think this will probably be a keeper.

Best of all, it’s free.

Google Desktop, I hear you say? Problems:

• Privacy – you may not be aware it can submit its search index of your personal files to a server. Superfluous lawsuit and subpoena? There went all your privacy. (Never mind that we have very little privacy by modern practices – unless extreme self-protection is your avocation.)
• Inefficiency; it is behemoth and sluggish because it indexes everything
• As a consequence of 2, it is mostly useless – when I have used it to search for email or file name text I know exists, it hasn’t found it – because it is still indexing the other 90% of useless information on my hard drive, and it hasn’t indexed what I’m looking for yet.
• No file tagging.
• I boycott Google when I can, for reasons I’ve blathered about here too often. Google it under this domain or search “google” in my blog’s (non-Google!) search tool.

Nope. tag2find wins hands down.

[Update 2010-01-31]

Just spotted this YouTube video about an upcoming “next generation” version of this tool. Watching this video, I’m completely baffled why an investor would drop funding for this. The project is apparently and unfortunately lagging for that reason. But watch this video.

They’re also planning an API that will allow external tools to interact with it, including, it seems implied, link tagging (like delicious.com) and web object tracking (such as individual photographs – and/or tags associated with them? – posted to Picasa)? If so, and from that video also, I think the next generation version will be very hot when it comes out.

From here.

The green bacterial looking clusters in the final sequence before it evolves into a blur – that is one of the coolest artistic abstractions I’ve ever seen. But I think the whole sequence is breathtaking.

For example – in Mozilla Thunderbird, to add an RSS feed and manage reading it in a sensible, organized way, you must follow detailed steps involving figures A through P (or 1 to 16; PLUS, if you want to know that this is 16 steps (simply to measure the extent of your technical fatigue), you must count A-P on your fingers. Unless you happen to know off-hand that P is the 16th letter of the English Alphabet. I didn’t.)

[The header of the concluding section of that article says: "That Was Painless". Um, if you're used to following, say, 42-step technical processes in your everyday work, maybe, in comparison.]

This should be a three to four step process:

1. Click the RSS icon in the address bar of Firefox.

2. Select “Subscribe Using Thunderbird”, which should be available by default (if you want Thunderbird to even be in the equation – which, even if it is, Firefox doesn’t tell you in the subscription button – it assumes you just know). Since this is not available by default, you have to follow a 5 step configuration to make that available.

3. Thunderbird comes up with a window, asking “If you want to add [title of RSS feed] to your Thunderbird RSS Subscriptions, select the account and folder you’d like to add it to.” – providing drop-down menus to select the account and folder.

Theoretically, it’s possible to get it down to such a three or four step task (and that page only got me heading in the general right direction). I haven’t gotten it to work. And this is in lieu of many more hacking steps I’d really rather avoid.

I thought these two applications were kinda cuddly friends? My exploration of the idea doesn’t find any proof..

Only a few months ago, researchers at Yale unveiled the first Electronic Quantum Processor.

It operates on two qubits, which exist in multiple states simultaneously (that’s the quantum mechanical aspect). When they add more qubits, they’ll be able to calculate multiples of multiple states in one processor cycle.

Excerpt:

Because of the counterintuitive laws of quantum mechanics, however, scientists can effectively place qubits in a “superposition” of multiple states at the same time, allowing for greater information storage and processing power.

For example, imagine having four phone numbers, including one for a friend, but not knowing which number belonged to that friend. You would typically have to try two to three numbers before you dialed the right one. A quantum processor, on the other hand, can find the right number in only one try.

“Instead of having to place a phone call to one number, then another number, you use quantum mechanics to speed up the process,” Schoelkopf said. “It’s like being able to place one phone call that simultaneously tests all four numbers, but only goes through to the right one.”

What is the potential? Here’s a way to spell it out mathematically, going off Wikipedia’s article on the topic:

A classical computer has a memory made up of bits, where each bit represents either a one or a zero. A quantum computer maintains a sequence of qubits. A single qubit can represent a one, a zero, or, crucially, any quantum superposition of these; moreover, a pair of qubits can be in any quantum superposition of 4 states, and three qubits in any superposition of 8. In general a quantum computer with n qubits can be in an arbitrary superposition of up to 2n different states simultaneously (this compares to a normal computer that can only be in one of these 2n states at any one time).

Where that describes a pair of qubits (two) in a superposition of 4 states, this means the qubits are in 4 different states at the same time. Following this, a trio of qubits (three) are in a superposition of 8, so that it follows the order of exponents or powers, which proceed like this:

• 2 qubits = 2 to the second power (2^2) = superposition of 4 simultaneous states
• 3 qubits = 2 to the third power (2^3) = superposition of 8
• 4 qubits = 2 to the fourth power (2^4) = superposition of 16
• 5 qubits = 2 to the fifth power (2^5) = superposition of 32..

With each additional qubit, the simultaneous states (or superpositions) doubles, so that:

• 8 qubits = 2 to the eighth power (2^8) = superposition of 256..
• 16 qubits = 2 to the sixteenth power (2^16) = superposition of 65,536..
• 32 qubits = 2 to the thirty-second (2^32) = superposition of 4,294,967,296..
• 64 qubits = 2 to the sixty-fourth (2^642) = superposition of 18,446,744,073,709,551,616..

What is that last extremely large number leading with an 18? That’s eighteen quintillion – going from thousands, to millions, to billions, to trillions, to quadrillions, to quintillions. More precisely, almost 18-and-a-half quintillion.

What does this all mean? Current computers operate in Gigahertz, meaning a billion calculations in one second; a computer processor with a speed of 3 Gigahertz runs around 3 billion calculations in one second.

(This is staggering, just by itself.)

When they create a sixty-four qubit quantum computer, it will be capable of running a calculation requiring around 18 and a half quintillion guesses in a few clock cycles (only a few millionths of a second).

Carl Sagan, eat your heart out.

Don’t get too excited yet. They haven’t figured out how to even build a computer around this yet. It’s only a processor.

But it’s a quantum processor. A two-bit quantum-processor, with quantum logic gates and a quantum bus.

With this kind of power, you’ll be able to find the 39-digit number which, when you run it through an image processing algorithm, will by algorithmic decompression happen to exactly match a digital image which without compression takes 1 gigabyte to store, but once you find the one out of 5 duodecillion 39-digit “fingerprint” numbers that match the image, you’ll be able to losslessly “compress” the image to only several hundred bytes. I don’t necessarily know what I’m really talking about here, but it will be something like that.

You live in a Star Trek universe.

One day, possibly in your future, this will look something like this article.

[Myeh. The Wordbook plugin copied the first draft with an innacuracy too soon at FaceBook, and I can't change it.]

I’ve become fascinated and very impressed by public-key encryption, which I’d never understood, but have now read up on at Wikipedia.

With this setup, you and you only hold a private decryption “key”. You also have a public encryption key. Anyone can encrypt anything with your public key, but anything so encrypted with your public key can only be decrypted with your private key. Your public key can therefore be completely public.
Read the rest of this entry »

As you know if you’ve been reading this blog, I’m a huge fan of these “Electric Sheep” images and the screensaver.  This morning I’ve started describing it in more detail at my Wiki page.

I’ve figured some basic ways to create my own original “children” Sheep by cross-breeding Sheep that someone else designed.  I’ve rendered them at a resolution to please virtually any computer “wallpaper” collector.  In the following two galleries, the link that says “open full image (click) ” is your friend

The first gallery is of the “parents” whose genes I crossed to create various children.  (Adobe Flash is required to view the galleries.)

The following gallery is of the “children” (original creations!) whom I thought were pretty.  (I killed the others.)  Some of these were found by panning, zooming, and scrolling through their loop animation with Apophysis.  Again the “open full image (click)” link is your friend.

Feel free to use and reuse these for any purpose.  The license is Creative Commons attrib. share-alike – and I request credit given to Richard Alexander Hall in reuse.  If you make derivative works from these, they’re completely yours.

I’d wait until I’ve added these to a page to market them as available for print on a huge poster (+ 2′ x 4′, like this one), but I haven’t the patience.  (I want to redesign that whole pick-a-sheep page as a blog page, anyway).  I’ve created some pretty things, and the world must know about it now!

Great news.  iTunes will drop copy protection on every song they sell.

If they also adopted LAME mp3 as their default encoder, added support for .ogg files to the iPod, reduced the software and memory footprint of iTunes by about half (it is such a hog!), and announced an iTunes extensibility/add-on API, I’d be wowed.

I use Winamp, but don’t really recommend it to people.  It’s a nerd’s music toy.  I use plugins with it that let me do a number of things:

• Rate songs, sort into cue folders (such as “edit”, “re-rip”, “audition”), send to recycle bin, etc. – using keyboard shortcuts (which really speeds up music tryouts)
• Auto-sort and rename files by my own custom preferences
• Rip from various video game music formats
• Play back a wide variety of music formats
• Navigate and cue music from a branching directory view
• Copy files anywhere merely by dragging them from the playlist
• Backup library and rating information
• Sync my collection (including “smart views” – which iTunes calls “smart playlists”, which for example will play all songs of genre x or y above a rating of 3, or everything with an “audition” comment in the tags, etc!) with my iPod.

That is all thanks to Winamp’s open plugin API, which invites the good will and genius of thousands of people who program so many plugins voluntarily.  iTunes can do only the last two on that list, and backing up the library never saved song ratings for me (that was several versions ago, so maybe they’ve clued into allowing that by now – I don’t know).

Winamp also has play/pause/rewind etc. keyboard shortcut capability “out of the box”.  It is also skinnable, meaning the way it looks can be very customized (thousands of skins have been created I’d guess), and optionally change at random with every played song.