Slashdot Mirror

The one that drives me bonkers isn't coding. It's Linear Algebra tests on paper, without even a four function calculator. Get one of dozens of arithmetic operations wrong, and the confused teacher tells the class /they/ are missing the forest for the trees.

In Linear Algebra, the important thing isn't doing dozens of arithmetic operations correctly - that's what computers are for. The important thing is knowing what formula or transformation to apply when, which pencil pushing just obfuscates.

When I took a Computer Graphics course, the one that talked about Beziers and Splines and stuff, we used a lot of Linear Algebra. I did my homeworks on paper, as requested, but I checked it with symmat, which I wrote just for that class.

For speed, it's hard to beat an rsync wrapper like Backup.rsync: http://stromberg.dnsalias.org/... . It doesn't deduplicate super-well, and it doesn't compress on disk, but like I said, it's fast. It's quite good at removing old data that you don't care about anymore.

For frugality, consider something like backshift: http://stromberg.dnsalias.org/... . It deduplicates quite well, and compresses everything (including most metadata) with xz. It also makes it easy to expire old data, unlike a lot of backup software you'll see.

Full disclosure: I wrote both of them.

Here's a table comparing some common backup tools: http://stromberg.dnsalias.org/...

For speed, it's hard to beat an rsync wrapper like Backup.rsync: http://stromberg.dnsalias.org/... . It doesn't deduplicate super-well, and it doesn't compress on disk, but like I said, it's fast. It's quite good at removing old data that you don't care about anymore.

For frugality, consider something like backshift: http://stromberg.dnsalias.org/... . It deduplicates quite well, and compresses everything (including most metadata) with xz. It also makes it easy to expire old data, unlike a lot of backup software you'll see.

Full disclosure: I wrote both of them.

Here's a table comparing some common backup tools: http://stromberg.dnsalias.org/...

For speed, it's hard to beat an rsync wrapper like Backup.rsync: http://stromberg.dnsalias.org/... . It doesn't deduplicate super-well, and it doesn't compress on disk, but like I said, it's fast. It's quite good at removing old data that you don't care about anymore.

For frugality, consider something like backshift: http://stromberg.dnsalias.org/... . It deduplicates quite well, and compresses everything (including most metadata) with xz. It also makes it easy to expire old data, unlike a lot of backup software you'll see.

Full disclosure: I wrote both of them.

Here's a table comparing some common backup tools: http://stromberg.dnsalias.org/...

http://stromberg.dnsalias.org/...

In four different programming languages: http://stromberg.dnsalias.org/...

You might be able to tell which hop is slow using something like pchar: http://stromberg.dnsalias.org/~strombrg/network-performance.html

I've commented on python and whitespace enough times, that it became more practical to create a small web page about it: Read it here.

Much as I love python, I see nothing to refute my objections to python's use of semantic spaces and lack of braces or other equivalent construct.

- It makes python ineffective as a shell-replacement language, because you can't easily do a complex one-liner and then retrieve it from history
- It makes cutting and pasting code difficult, which makes refactoring unnecessarily painful
- It makes it impossible to fully auto-format code: all formatters I have seen for python do not trust themselves to touch leading spaces, nor should they. This leads to ugly, inconsistent-looking code.

It's not just about the obvious problem of tripping up anyone who is not paying attention to this peculiarity of the language. If python had semantic line ends (but not semantic leading spaces) with optional semi-colons (for when you don't use a line end) plus braces, the amount of extra typing and visual clutter would be minimal, while these problems would go away, making python a more useful language.

Here's a list. See especially CFFI.

Python, as a plugin, would require adoption by Chrome, Firefox, Internet Explorer, Opera - I doubt all four of these would all agree to support Python as a plugin. However, it's possible to compile Python to javascript. Most of these transpile individual Python programs to JavaScript, but one, empythoned, actually compiles CPython 2.x to JavaScript using LLVM's JavaScript backend.

I've commented on python and whitespace enough times, that it became more practical to create a small web page about it: Read it here.

I've put a deduplication technology comparison table here.

Most deduplication technologies are SIS (deduplicates only entire file content at a time - requires O(n^2) storage for large, slow growing files) or fixed-width blocks (has problems if you change a byte at the beginning or middle of a file).

The strongest offerings do sliding block (or what I've been calling variable-length, content-based blocking). These don't have problems with inserting or changing or deleting a byte somewhere in a file.

I've designed and coded a backup system that does variable-length, content-based blocking for deduplication, called Backshift. It's very nearly ready to hit 1.0, mostly just needing a few more users to try it out. It's got a comprehensive automated test suite, lots of documentation, runs on all the major Pythons except IronPython (fastest on Pypy - Pypy's even faster than CPython combined with Cython for this application), and has been ported to a variety of Linuxes, DragonflyBSD, FreeBSD, OS/X, Cygwin, Haiku, Solaris and Open Indiana.

In addition to the deduplication, it compresses the deduplicated chunks with xz (with a bzip2 fallback if none of the 3 methods of doing xz work). Also, if it notices that a chunk grows when compressed, that chunk gets stored uncompressed - so backing up a file that was already compressed pretty hard doesn't require increased storage in the backup repository.

Use:

• To backup a filesystem, you pipe find -print0 to it.
• To restore a directory, you give it a starting directory and ask it to assemble a tar archive from the pieces and pipe that to GNU tar (which of course can optionally be piped over ssh first for clientless remote restores).
• To verify a backup, you can easily use GNU tar's verification mode and a file count.
• To list what's in a backup, you give it an optional starting directory and ask it to list the files.

The subdirectory specifications are excelerated quite a bit over what plain tar would give.

It sometimes acts so much like tar (especially for restores and listing contents) that you might be tempted to think that it's storing things as tar archives behind the scenes - but it's not.

I've put a deduplication technology comparison table here.

Most deduplication technologies are SIS (deduplicates only entire file content at a time - requires O(n^2) storage for large, slow growing files) or fixed-width blocks (has problems if you change a byte at the beginning or middle of a file).

The strongest offerings do sliding block (or what I've been calling variable-length, content-based blocking). These don't have problems with inserting or changing or deleting a byte somewhere in a file.

I've designed and coded a backup system that does variable-length, content-based blocking for deduplication, called Backshift. It's very nearly ready to hit 1.0, mostly just needing a few more users to try it out. It's got a comprehensive automated test suite, lots of documentation, runs on all the major Pythons except IronPython (fastest on Pypy - Pypy's even faster than CPython combined with Cython for this application), and has been ported to a variety of Linuxes, DragonflyBSD, FreeBSD, OS/X, Cygwin, Haiku, Solaris and Open Indiana.

In addition to the deduplication, it compresses the deduplicated chunks with xz (with a bzip2 fallback if none of the 3 methods of doing xz work). Also, if it notices that a chunk grows when compressed, that chunk gets stored uncompressed - so backing up a file that was already compressed pretty hard doesn't require increased storage in the backup repository.

Use:

• To backup a filesystem, you pipe find -print0 to it.
• To restore a directory, you give it a starting directory and ask it to assemble a tar archive from the pieces and pipe that to GNU tar (which of course can optionally be piped over ssh first for clientless remote restores).
• To verify a backup, you can easily use GNU tar's verification mode and a file count.
• To list what's in a backup, you give it an optional starting directory and ask it to list the files.

The subdirectory specifications are excelerated quite a bit over what plain tar would give.

It sometimes acts so much like tar (especially for restores and listing contents) that you might be tempted to think that it's storing things as tar archives behind the scenes - but it's not.

Like I've explained in a previous post in much greater detail, the GPS of a Nokia phone (even with its free off-line Ovi vector maps) is almost completely useless without a data connection.

Then you have been misleading people.

I use my Nokia N85 when I go walking and it works very well. It doesn't even need a SIM card to be useful, let alone a data connection.

Combined with Trek Buddy and maps downloaded with Mobile Atlas Creator it is a competent alternative to a dedicated GPS unit.

However, no phone is as robust or waterproof as a purpose built device. Battery capacity is also a concern, though this can be alleviated by carrying chargers such as the Nexus Poerboost which can be recharged via USB or use standard AA batteries.

It should also be remembered that a GPS is a navigation aid, and does not negate the need for a map and compass, and the ability to use them.

I drive the I-10 through West Texas, and see a huge (250-foot?) cell tower every mile or so.

But it's worth fuck-all to me if I can't get Google Maps to work because they refuse to put new equipment on those towers to handle data services.

There are two ways to go about this.

If you want a "just works" solution, go and buy CoPilot for your smartphone (iPhone/Android). It's your run-of-the-mill GPS navigation software, with UI like the hardware units, and offline maps. The version with North America maps is $30.

If you don't mind wasting time, you can get any of several apps that can display preloaded offline maps. For Android, I use Maverick. Then you take something like Mobile Atlas Creator, which downloads Google (Bing, ...) maps on your PC and prepackages them for offline use, and upload them to the phone. Voila! Though, given the effort, I mostly use this approach when going hiking - preloading the particular region with terrain and satellite layers in highest detail.

I will have to disagree this is solely a human problem. Technology can solve this. But it is not cheap. It is claimed that a better to give a rough location rather than to triangulate, which requires a subsequent mapping. Details can be found in K. K. Yap, V. Srinivasan, and M. Motani, MAX: Human-Centric Search of the Physical World, Proceedings of ACM Sensys 2005, San Diego, CA, USA, November 2005. Available at http://wine.dnsalias.org/motani/publications.php But the cost is quite forbiddingly now. I actually have a working prototype but it is nowhere near economical. Good luck and let me know if you can find passive cheap readers with the range needed (about 1m). Disclaimer: I authored this paper.

By: Chris Carlson of http://boog.dnsalias.org/chris/

Several major Venezuelan journalists have received all-expenses paid trips to the U.S. for courses in an apparent effort of the U.S. State Department to influence the media in Venezuela, according to recently released documents. The Venezuelan-American attorney Eva Golinger, who released the information yesterday in a press conference in Caracas, also revealed evidence of a destabilization plan against the Chavez government to take place this weekend.

Golinger is the author of The Chavez Code, which documents U.S. funding of opposition groups and U.S. involvement in the 2002 coup attempt.

Under a program named International Business Leadership Program, many Venezuelan journalists, mostly from the opposition media, but also some from the Venezuelan government, have received "scholarships" from the U.S. government to attend training courses during the years 2001-2005.

Some of the most recognized opposition journalists of the country have participated according to the documents, including Miguel Angel Rodriguez of RCTV, who received more than six thousand dollars for his participation in 2003, and Maria Fernanda Flores of Globovision among others, according to the documents obtained by Eva Golinger through the U.S. Freedom of Information Act.

With the supposed intention of teaching journalists about the media and journalism in the United States, the program also has the purpose of influencing how Venezuelan journalists cover events related to the U.S. foreign policy. According to the documents released, the programs denominated "Journalism IV" seek to "influence the approach and ultimately the coverage given to issues of importance to U.S. foreign policy and to strengthen the Venezuelan democratic process."

The State Department gave special attention to the Venezuelan news channel Globovisión, which they believe to be "the most influential channel" and to have the most positive coverage of the United States. The State Department sought a special relationship with this particular news network, and especially with one important journalist Maria Fernanda Flores.

According to an unclassified State Department memo, "A program that gives Flores a better understanding of and closer ties with U.S. media executive decision-making policies and practices can help Globovision, already the country's news leader, an even more professional responsible force in Venezuela's media environment, with profound implications not only for more positive coverage of U.S. policies but for Venezuela's evolving political situation as well."

Golinger emphasized, though, that the journalists involved in these programs were chosen by the U.S. embassy and could very well be unaware of the program's efforts to influence their coverage of U.S. foreign policy.

Golinger also spoke about other State Department programs including one to increase U.S. access to the Venezuelan Armed Forces through various training programs, whose objectives she said are similar to the program for journalists.

The Press Attaché of the U.S. Embassy in Venezuela, Bryan Penn, responded on Globovisión to Golinger's press conference yesterday by saying that the programs she presented were common with governments around the world and that the U.S. is "proud of them."

Destabilization Plan

Golinger also presented evidence of a destabilization plan for this Saturday, showing a flyer calling for people to come into the streets and march in the morning hours of Saturday, May 26th. According to the attorney, the campaign is designed by Freedom House, a U.S. organization dedicated to non-violent resistance.

Freedom House, headed by Peter Ackerman, has been involved in other countries and other campaigns to overthrow regimes such as Serbia and the Ukraine. According to Golinger, the flyers circulating in Caracas have the logo of a clenched fist, the same logo used in the campaigns in other c

So, I had to try this out and see it for myself. And, sure enough it rendered correctly. Then I started noticing the problems.

Here is a screenshot after "scrolling" using either the scroll wheel or up/down keys (despite the fact — as you point out — that there are no scroll bars).

And another one after a resize of the window. I restarted konq before doing the resize, so the issues aren't left over from the scrolling.

Also, note in the resized screenshot that the progress bar is stuck at 37%.

So, IMO, close, but not quite. However, close is better than most of the other browsers out there.

So, I had to try this out and see it for myself. And, sure enough it rendered correctly. Then I started noticing the problems.

Here is a screenshot after "scrolling" using either the scroll wheel or up/down keys (despite the fact — as you point out — that there are no scroll bars).

And another one after a resize of the window. I restarted konq before doing the resize, so the issues aren't left over from the scrolling.

Also, note in the resized screenshot that the progress bar is stuck at 37%.

So, IMO, close, but not quite. However, close is better than most of the other browsers out there.

who loves BSD! I still do!

in other news pkgsrc on SFU updated, the new pf from obsd/3.8 is getting ported, there is also a kernel emulator for fbsd/5+ and a smbus implementation. matlab works too and some people might be interested in a list of translations for the `of course it runs netbsd' motto.

what else can be said, its great that *the* bsd is still alive

Now let's get the Federal, State, and local governments to create standards-based web sites as policy. (Remember FEMA and the US Copyright Office.)

http://narnia.dnsalias.org/freegovernment/

Next, let's go for standards for electronic office documents (word processor, spreadsheet). I doubt Lockheed Martin's system here is going to understand every obscure format---say Microsoft Word 5?

You can mostly cut and paste this letter for sending to your local government leaders regarding web standards.

http://narnia.dnsalias.org/freegovernment/

One way or another, politely bring up the issue and mention the benefits from the government's point of view: serving people, long-term savings, etc.

There might be a law if enough concerned citizens write their government leaders. Here is a letter template for this very issue. Copy, paste, edit, send, and repeat.

http://narnia.dnsalias.org/freegovernment/

The site also includes some research and related links.

We need government policies that require a vendor-neutral web.

In addition to the Baen free library, Baen has also put 7 freely distributable cds containg many books into some of their hardcover books, with the text "This disk and its contents may be copied and shared but NOT sold."

This site and this site both host the first 6 cds as web browsable books.

This site contains torrents for all 7 cds.

What a great photo Tom. Captured the moment very well.

Shouldn't that be BIG RED SHINY BUTTON?"

p.s.
lameness filter encountered: don't use so many caps.

-

After all, both parties are cut from the same cloth.

I'm amazed that anyone who believes that has the brain power to keep their lungs functioning, because of their overwhelming stupidity. You sound just like Nader, for saying there was no difference between Bush and Gore. The common clay of the new West. You know, morons.

Here ya go...be nice, OOL gives me a lotta upstream bandwidth (yes, shameless plug) so let's try and not give them a heart attack, mmmkay?

check it out..
..although i cant see iterm being too useful like that..

Thanks for pointing that out. My confession is archived here.

But, myself I used Qbasic the other day to take some data and generate a formated HTML file with it...

I will probably regret linking to my 56K from Slashdot, though.

well, at least the closeup of a sunspot and one of the filaments. but please be nice, it's a new powermac, i don't want it melted just yet :P

well, at least the closeup of a sunspot and one of the filaments. but please be nice, it's a new powermac, i don't want it melted just yet :P