| Abram Hindle's Blog | |||||
|
Subscribe
Flavours
Links |
Sun, 22 Mar 2009
Sketcha-mo-phone
I've made an instrument using my arduino, some LEDs, some resistors, a breadbord, some wires, and a pencil. The pencil provides us with a kind of potentiometer. This allows us to use a line draw by hand, much like a guitar string. Thu, 12 Mar 2009
Qi II is NOT Open Source at all, M. Tarver is abusing the term open source with
It really bothers me that educated people would take part in the dilution of semantic meaning of something so legalistic as licenses. To: dr mtarver ukonline co uk From: abram hindle Subject: Qi II License Your license for Qi II is really confusing, it says "not for the production of commercial software", commercial software is not defined, then later it says "remain available as open source". If you read http://www.opensource.org/docs/osd , which is considered to be the standard for the definition of open source (since you did not define it) Clause 1 states: 1. Free Redistribution The license shall not restrict any party from selling or giving away the software as a component of an aggregate software distribution containing programs from several different sources. The license shall not require a royalty or other fee for such sale. Clause 1 means that I can package any such open source software, put it on a CD, sell the CD, charge money for any services related the product etc. Even in a commercial setting. This is required of opensource, it shouldn't matter how its used as long as the license and property is respected (which can require that all future derivatives remain under the same license). So your license tries to be open-source but contains clauses which violate the spirit open source. MySQL gets around this by dual licensing and requiring copyright assignment in the case of accepting patches. Can you address these issues?
If it was dual licensed with the GPL and whatever license you want then at least I'd know it was opensource, but as it stands Debian can't include your software with their distribution and it certainly doesn't sound like it is open source. abram License found at: http://www.lambdassociates.org/whatsnew.htm This software is licensed only for personal and educational use and not for the production of commercial software. Modifications to this program are allowed but the resulting source must be annotated to indicate the nature of and the author of these changes. Any modified source is bound by this licence and must remain available as open source under the same conditions it was supplied and with this licence at the top. This software is supplied AS IS without any warranty. In no way shall ; Mark Tarver or Lambda Associates be held liable for any damages resulting from the use of this program. The terms of these conditions remain binding unless the individual holds a valid license to use Qi commercially. This license is found in the final page of 'Functional Programming in Qi'. In that event the terms of that license apply to the license holder. And Dr. M Tarver responds: From: Mark Tarver To: Abram Hindle Subject: Re: Qi II License I don't adhere to the open source definition of Debian which is not trademarked nor binding. The source you quote to a retrospetive attempt to bind the term. For me, it means that the source code can be viewed and changed. Commercial = for money or anything of monetary value. Wed, 10 Oct 2007
Subject: IP Forwarding on OpenVZ for OpenVPN?
Read this thread: http://forums.vpslink.com/showthread.php?t=1864&page=2
Subject: Corrupt countries were more likely to support the OOXML document format? With a correlation coeffecient of -0.31?
http://www.effi.org/blog/kai-2007-09-05.en.html For some reason they use a Wilcoxian test and fisher t test on the data. The data has 2 random variables, corruption (continuous) and vote (yes/no) They said they should with good p-values that there correlation between voting for approval and corruption. Well using both Pearson and Point-Biserial Correlation coeffecient I found a coeffecient of -0.3129 with a p-value of 0.009373 this means there was a weak or medium stength linear correlation between corruption (more negative) and approval votes.
What's interesting is that EFFI keeps applying the wrong tests when they
want to show to correlation. I also did the rank based correlations and
they weren't as good as the linear cases -0.29 for Spearman-Rho and 0.24
My data and a log of what I did is at: http://churchturing.org/w/ooxml abram -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFG4KGJnOrfa1yW8IURAhzJAJoCHDfl/vzKWrdoMNK3XrJQxpIaqgCfX0Ci UwcoNDr9n+tSnybrRJ3wcWI= =Ob5D -----END PGP SIGNATURE-----
Subject: OCaml JACK JACKit Support
To get it run:
Subject: NVidia on Feisty or Gutsy
Anyways read through http://doc.gwos.org/index.php/Latest_nvidia_feisty for instructions on how to handle your card. abram
Subject: Hon. Maxime Bernier - Copyright Act
Pretty bland response. I editted some email addresses.
---------- Forwarded message ----------
Date: Wed, 30 May 2007 10:15:26 -0400
From: "Correspondence Minister/Correspondance Ministre: OCS"
Thank you for your electronic correspondence regarding possible amendments to the Copyright Act (the Act). In my view, the Act must continue to be supportive of innovation and research while reflecting current technological and legal realities. To this end, a balance between adequate protection for copyright holders and reasonable access to copyrighted material is critical. With this in mind, I am working closely with my colleague, the Honourable Bev Oda, Minister of Canadian Heritage, to determine the appropriate next steps with respect to copyright reform. Please be assured that your comments will be taken into account as we move forward. Sincerely, Maxime Bernier -----Original Message----- From: Abram Hindle [mailto:ahindle cs uwaterloo ca] Sent: April 16, 2007 9:12 PM To: Correspondence Minister/Correspondance Ministre: OCS Subject: Proposed changes to BILL C-60 Dear Maxime Bernier, Minister of Industry, As a computer science researcher I am concerned over news that anti-digital-circumvention laws are proposed as additions to BILL C-60. By making digital circumvention illegal and the tools and means of doing so illegal you provide NO market incentive to industry to protect a user's privacy. Essentially computer criminals will use the latest software exploit to access the private information of Canadians and use this info in expensive and dangerous ways. They will do this regardless of legality of circumvention. What this law does is it silences ethical researchers who find bugs in software and devices which often protect people. Without these researchers these bugs go un-found or they are kept within the computer underground and used against unwitting victims. Essentially you're taking the most capable researchers who help protect your digital security and putting them in chains. Even worse you're allow industry and the market to become lazy, you're removing incentive for industry to fix security issues. If you think these are laws are only for music and videos I suggest you read about the abuse of the DMCA in the USA and how it is used to protect faulty voting machines and buggy software from "exploitation" by people who have every right to audit the safety of these devices. Do not make common sense illegal, software research has shown that there are numerous bugs created per every 10 lines of code written. Most software packages you use are over 100,000 lines of code. Bugs are not going away and computer exploits are not going away. Do not disable our ability to secure ourselves against criminals. Ref: http://www.michaelgeist.ca/content/view/1875/125/ Abram Hindle
Subject: Ric's brilliant idea
Ric then came up with the idea of getting the computer to read your work to you. The computer reads accurately and perfectly, it only leaves out things it doesn't understand. It doesn't skip over double words like "the the" and doesn't say "blah blah" when a word is too long. Also, its tongue never tires. Anyways, in a fit of yak shaving I made an emacs macro that piped text from a region to festival (a CMU text2speech engine). I found that emacs blocked even if I ran festival with "&" to run in the background. So I made a daemonize script to run festival (I copied it from perldocs). So now when I hit ESC-\ on a region in emacs it pipes the text through untex and into festival, which then reads to text outloud to me using a Scottish voice. Was it helpful? Yes it was. I felt that it helped immensely. It would read text accurately and not skip words, a problem that I have. It wouldn't tire me out. It didn't hurt my eyes, I could just follow along slowly. The benefits of a speech synth is that it makes it apparent when your timing is off, when you are missing a comma. When a sentence is too long the voice runs out of breath. When a sentence keeps repeating similar sounding words too much you will notice it. Even more interesting is that it seems to use a different part of the brain to process what has been written. When you listen to something that is incorrect you can usually notice it. Anyways, it made editing a lot more fun and I felt more confident about it. Perhaps it was novelty, it seemed to work. If I wanted to check if what I wrote was correct I just made it read it out and it became pretty apparent if I was missing something. I've inlined the results of the yak shaving below (it didn't take a lot of time). I did a tiny bit of research and it seems some people advocate this, and some people even sell software to do this. Thanks for the great idea. abram Attachments:
The macro in .emacs :
texspeak:
festivaltts:
Subject: KDE is coming along quite well
You can use dcop from the commandline to automate various KDE related tasks. This is the kind of desktop integration we need more of. The ability to use graphical apps yet still script them. This kind of thing has been available for a while but is really just starting to really catch on in KDE and GNOME on linux. It has been available to Windows for ever and OSX has their appletalk and automator interfaces.
Subject: Good MPI article
Subject: Blurring
I wouldn't have gone about this problem the way he did. I would've used feature vectors and machine learning to characterize and classify the bitmaps (which he'll probably have to do). Though what I thought was great is he actually analyzed the problem and came up with a very novel solution (I doubt it works in all cases but still close enough). Sun, 10 Sep 2006
Subject: Latest Movie Recommendation
The Caulkins plot to teach a bully a lesson. Well acted and serious. One of the few movies with kids in it which is actually any good. Wed, 30 Aug 2006
Subject: Gmail admins need to get their act together
Received-SPF: softfail (gmail.com: domain of transitioning iywhk@abez.ca does not designate 63.229.116.57 as permitted sender) Go look up SPF, it is a way of denying hosts from sending mail as you by allowing spam filters to filter based on a txt entry for your domain. Regardless on a softfailure gmail is sending me spam from people pretending to send mail from my domain. I've sent them emails complaining about this in the past but they've been ignored.
Subject: Multiple Keyboards and Linux
I've been recently trying to get multiple keyboards providing me multiple input. There are a few ways one can do this:
Everytime you'll have to parse the event devices. Good things to search for in google to help you out "multi seat linux". You might need a ruby kernel. I'm not done yet so I think I'm going to hack usbkdb and tell it to ignore the USB keyboards I have. I have not verified that it'll still work as an event device (I sure hope so). abram Mon, 28 Aug 2006
Subject: Even more china photos
Wedding Photos:
Sunday outing:
Subject: More Photos
Here are photos of some dinner & hong kong: http://churchturing.org/w/20060826/ http://churchturing.org/w/20060827/ I liked hong kong because it was cleaner although it was more expensive than the mainland. abram Tue, 22 Aug 2006
Subject: China
Here are some pictures: http://churchturing.org/w/2006-08-15-2214-03/ http://churchturing.org/w/2006-08-15-2217-50/ http://churchturing.org/w/2006-08-15-2237-50/ http://churchturing.org/w/20060822/ Fri, 28 Jul 2006
Subject: Marital Status
Formal Cermony will be held August 19th in ShenZhen, China. I'll have to wear something strange. Tue, 25 Jul 2006
Subject: DJBDNS (DNSCACHE)
Subject: New Abez Font
http://churchturing.org/w/abez-0.2/
Here's a new abez font and it works a little better:
X11: http://churchturing.org/w/abez-0.2/abez.pcf windows: http://churchturing.org/w/abez-0.2/AbezMedium-0.2.ttf OSX: http://churchturing.org/w/abez-0.2/AbezMedium-0.2.mac.dfont debian/ubuntu: http://churchturing.org/w/abez-0.2/xfonts-abez_0.2_all.deb Thanks to mikael http://www.deepwood.net/~mikael/tumblelog/ for the deb and Makefile (which I probably broke). Oh the purpose of the font is to have a thin font which is about 13-15 pixels high as well as do nice line art. Mikael made the braces and parans and brackets very pro-fontish. abram Thu, 08 Jun 2006
Subject: Social Networking
http://www.newscientist.com/article/mg19025556.200?DCMP=NLC-nletter&nsre... Mon, 05 Jun 2006
Subject: PrintfGentle
A Type Safe Printf in MLTon. Impressive. Wed, 03 May 2006
Subject: More Chicago Photos
Subject: Chicago Stuff
http://churchturing.org/w/20060501/ http://churchturing.org/w/20060430/ YEAH INTERNET FRIENDS. Sun, 30 Apr 2006
Subject: Chicago Pics
http://churchturing.org/w/20060430/ IO_Burn wants gr33ts. So gr33ts 2 IO_BURN. Fri, 28 Apr 2006
Subject: In Chicago
Subject: Oh wow GPG
http://abez.ca/~abez/public-key.asc <--- here's my public key if you need it. (for abez@abez.ca) Wed, 08 Mar 2006
Subject: Random Entry
Best of all, I could this wearing a wizard cap. abram Tue, 31 Jan 2006
Subject: Casper Was Helpful
http://churchturing.org/w/AbezMedium.ttf
Subject: Made a new font
http://churchturing.org/w/misc-abez.bdf
http://churchturing.org/w/misc-abez.pcf
Subject: WINE
abram Sun, 15 Jan 2006
Subject: Speech Recognition On Linux
I've tried different packages from CMU Sphinx2 Sphinx-2 Sphinx3 Sphinx-3 Sphinx4 Sphinx-4 I first tried Sphinx-4, it worked well for live demos but when I tried to integrate more complex languages into it, such as HUB4 or WSJ I just couldn't. Classpath problems and like even though I directly referenced both the jar files and class files. So Sphinx-4 was a bust simply do to it's very lame and over XMLify configuration system. Making if very difficult to get started. Then I started on Sphinx2, I tried to make my own LM files from random internet text and that was very troublesome. I could only use their web service to generate the files from text files. They only allowed for 5000 uniq words as well. Once I had that it was easy to run sphinx2-demo with the new ml and dic files (put into a new directory) except that it was very very picky of what I said. So picky that rarely did it ever print any output at all. Very disappointed I tried to figure out how to make it more liberal. I could not find the right parameters at all. So I was limited by vocabulary sizes, I was limited by the transitions between words in these vocabularies, and I was limited by the strict sensitivity of sphinx-2. I was rather unhappy. I had searched for general models etc and couldn't find them. Until I remembered, with Sphinx 4 I had tried to use HUB4 and WSJ data... So I tried to the HUB4 dump file with sphinx2 (to no avail). I then searched for hub4 and sphinx2... I found: http://www.arborius.net/~jphekman/sphinx/full/index.html So now I am waiting for this guarantuan file to load and see if it will recognize anything useful at all. Sadly it doesn't work well for general dictation at all :( abram Tue, 10 Jan 2006
Subject: I wasted time time solving sudokus
Getting the computer to solve them in perl :| Runs reasonably fast. Mon, 26 Dec 2005
Subject: Christmas letter
Lixin and I are settling into our new home in Waterloo. We live right downtown, so we have access to all the amenities and Lixin can go to work by bus. Lixin has found work as a private tutor and a Supply Educational Assistant at Waterloo School Board, this means she gets real classroom experience. She went to different schools (primary or secondary) almost everyday. Thus she saw various ways/styles of teaching and learning. She is having fun and is very busy. I, Abram, have been exceptionally busy with a full course of 3 graduate courses while acting as a teaching assistant for a course he calls "Programming for Newbies". The courses were "Software Architecture", "Artifical Intelligence: Reasoning through Uncertainity" and "Databases: Web Data Integration and Stream Databases". The database course required at least 15 hours of work a week. This combined with random 16 hour marking stints virtually killed me. Unfortunately one course is left. The professor said the final report is due in January. Anyways the courses at Waterloo are more work intensive than at UVic, I'm not sure if they are harder but there is more work. I think it is more busy work to be serious. The course work has impeded my ability to think. Thankfully Lixin is around. What I do know is that Ontario is expensive, dirty and busy. Waterloo feels like a good school. Waterloo has lots of good students who care about research. I have made some friends who enjoy UNIX and IRC and other computer stuff I enjoy as well. We hang out in a channel #beer and meet up at the Grad House for dinner or so they can drink (I drink Coke). Christmas is consists of Lixin and I in Waterloo. We spent the afternoon of Christmas Eve cleaning and recycling. After that, we went to the mall which promptly closed on us, then we had a great meal at the Keg, followed by the Narnia movie which was entertaining. We came home and cuddled up watching the Aviator while Abram was chatting online at the same time using his new laptop. I, Abram, look forward to the next term where I take a course about how to generate and draw flower patterns onto spheres. My TA duty will be Tech Support to both students and faculty (at least it is regular hours). Anyways I expect the next term to consist of less work I don't care about and more actual research. http://churchturing.org/w/postgresql.swf is an example of some of the fun I had this term. Anyways I'm doing better now, adjusting to Ontario, Waterloo etc. I hope to get some papers out this coming term and probably go to China for the MSR workshop.
Love,
Subject: :|
Weather wise: More snow but it is scraped off the roads quickly. It is slippery. I'm not coming home for Christmas :(, it'd cost too much, I'll probably spend some of the time slacking off and watching Dr Who. Thu, 17 Nov 2005
Subject: White Stuff?
Subject: Worthless
I've met some new friends here at Waterloo, thank god they use IRC otherwise I'd never talk to anyone other than Lixin. Lixin seems to be doing ok, she's trying to get an job in education. I'll try to update more as it doesn't take that long. Thu, 13 Oct 2005
Subject: LONG TIME
Here's the latest picture set, it includes across Canada as well as Waterloo. http://churchturing.org/w/20051013 Tue, 21 Jun 2005
Subject: Saddened by Bram Cohen
Tis sad that Cohen doesn't quite get the benefits of using something like Reed Solomon codes with Bittorrent. The primary benefit is to reduce the rarity of blocks and to allow for an effecient resolution of the 99% problem where peers are searching the network for the last block. You might ask how does this reduce rarity? Well imagine our seeder is spitting out blocks in a linear order, after it is done the data blocks it starts to spit out error code blocks. Basically a leecher just need to wait til he has enough data and or code blocks to rebuild the file (the size of the file). If these error blocks are in the network other leechers can grab them and once enough are accumulated rebuild their file. This reduces rarity because each error block is not localized, it is related to the whole, so it doesn't really matter what combination we use. (I'm not saying avalanche uses this). The point is that suddenly we have a bunch of blocks which by working together act as other data blocks, then effectively reduces the rarity of data because you no longer need to ask for just that one block, you can be taking code blocks while you are searching for relevant data blocks. So you send code blocks out when there is extra bandwidth, but from the seeder point of view it doesn't matter which blocks they send out, if they send out code tho it allows people to rebuild easier. We are effectively adding more blocks to the network almost increasing the rarity of other blocks but this counteracted by the fact the error code blocks are not locallized and can rebuild other blocks. Mon, 20 Jun 2005
Subject: Go Daddy guy Bob Parsons admits belief in Fascism
With praise for the Cuban based US military base which is mired in scandal over torture, Bob Parsons comes out in support of it. Even worse when posters post facts about the US's involvement in past events and abroad he says he simply does not believe it. Bob, I guess you're allowed your beliefs but you can't refute facts just by not believing them. If that were true I'd be cruising throught the Grand Canyon at 500 km/h with my new anti-gravity beliefs. abram |
||||