I use TeXML in production, it’s an important part of our new publishing process.

Why the project is inactive? Because no bugs are reported for the version 2.0.1, released on 19 July 2006 (today is 10 April 2008). If it works well, why bother to change anything?

As a temporary solution, I found how to re-connect from a command line, thanks to the blog article “airport - the Little Known Command Line Wireless Utility”. I’ve created the file reconnect.sh and run it when required:

sudo airport -ASID –password=’$WEP_key‘

“How to create a symbolic link under Win2K, WinXP, and above?” — this google-answers document was my starting point.

My notes:

CreateHardLink function establishes a hard link between an existing file and a new file. This function is only supported on the NTFS file system, and only for files, not directories.

CreateSymbolicLink function creates a symbolic link. Requires Windows Vista or Windows Server 2008.

And finally, junctions or reparse points (from http://msdn.microsoft.com/library/default.asp?url=/library/en-us/fileio/fs/reparse_points.asp):

A file or directory can contain a reparse point, which is a collection of user-defined data. The format of this data is understood by the application which stores the data, and a file system filter, which you install to interpret the data and process the file.

This feature can be used to create links to directories. It is exactly what I need!

The tool to exploit the feature: Junction v1.05.

The straightforward way — convert to PostScript and the rest is known — doesn’t work:

$ djvups book.djvu book.ps
$ psnup2.pl book.ps >book2.ps

Unfortunately, gv can’t read the output PostScript file. The tool psnup works correctly only for a subset of PostScript, and djvups generates the code outside this subset.

An usual workaround is to use ps2pdf and pdftops — we first convert PostScript to PDF and then convert back from PDF to PostScript. But in this case, the workaround works only partially. The output PostScript is very, very big, and gv needs eternity to load it.

After trials and errors, I decided to try TIFF as an intermediate format. And it is the solution.

Well, the simplest way again doesn’t work:

$ ddjvu -format=tiff book.djvu book.tiff
Core dump

Instead, I extract page-by-page, ignoring the coredumped ones (fortunately, these pages are not important).

Step 1: get the number of pages

$ djvudump book.djvu | head
...
    DIRM [2733]       Document directory (bundled, 354 files 350 pages)
…

Step 2: write the page extraction script gen.pl.

foreach my $i (1...350) {
  printf "ddjvu -format=tiff --page=%i book.djvu pages/p%03i.tif\n", $i, $i;
}

Step 3: extract pages

$ mkdir pages
$ perl gen.pl >x.sh
$ sh x.sh
.. ignore core dumps ...

Step 4: create PostScript from TIFF

$ tiff2ps -w 8.27 -h 11.69 -c pages/p???.tif >print.ps

Step 5: make two pages per sheet:

$ psnup2.pl print.ps >print2.ps

Step 6: print!

Unfortunately, the trackbacks are extremely vulnerable to spamming. Indeed, trackbacks must work without human intervention (by design). Automatic filters doesn’t work well for blogs. And adding manual force breaks the design of trackbacks.

Fortunately, there is a simple solution.

We have blog aggregators and blog search engines. These services know about each blog on the planet and they can

1) distinguish the spam blogs from the good blogs,
2) find all the blogs which links to my blog post.

Therefore, a tool can search for the new links to my blog and create trackbacks records in my blog.

Other variations are possible as the main idea is fruitful:

Trackbacks should be performed through an intermediate, not directly.

P.S. Waiting for a such tool, I switched off the trackbacks for this blog. But commenting still work.

Here is how I feed the games to the computer.

Step 1: get the game

In FICS console, say (to get the games in the PGN format)

fics% set pgn 1

To get the list of recetly played games:

fics% history olpa

In the output, the first number is the game number. To get the game by e-mail, say:

fics% mailstored olpa game_number

Step 2: analyze

Just use your favourite chess program. I use crafty.

At the moment, I use the command:

annotate game.pgn wb x-999 .1 10

Details are in the documentation.

After annotating is finished, I use gvim or XBoard to see crafty comments:

$ xboard -lgf game.pgn.can

The Mysql database should be created this way:

CREATE DATABASE name CHARACTER SET utf8 COLLATE utf8_general_ci;

The file wp-config.php should contain:

define('DB_CHARSET','utf8');

In case of troubles, the database must be revised first. Check the UTF8 options and texts using phpMyAdmin.

0. Plug the microphone jack into the socket on the back side, not on the front panel.

1. Load all the required drivers

# modprobe snd-hda-intel
# modprobe snd_pcm_oss
# modprobe snd_seq_oss

1b. Setup mixer. (aumix)

2. Record

arecord -f cd test.wav

3. Convert to mp3

ffmpeg -i test.wav test.mp3

>>>  'foo'[::-1]
'oof'

Finally, I’ve found a working sequence of conversion commands.

cp latex/312nn081.tex tmp/312nn081.tex

cd tmp; latex 312nn081.tex

dvips -E -o tmp/312nn081.eps tmp/312nn081.dvi

pstoedit -f ps -dt tmp/312nn081.eps tmp/312nn081.ps

ps2epsi tmp/312nn081.ps tmp/312nn081_2.eps

../pictures_converter/bin/epstosmth --gsdev=tifflzw \
       tmp/312nn081.eps --outfile=tmp/312nn081.tif

epstool --add-user-preview tmp/312nn081.tif tmp/312nn081_2.eps \
       eps/312nn081.eps

I already wrote about “epstosmth”: http://uucode.com/blog/2005/09/29/eps-to-something/

The main feature of EPS files produced this way:

Both Adobe Illustrator and Adobe FrameMaker open the EPS files without complaints.

InPlace CMS
http://inplace.sourceforge.net/

Use InPlace CMS to maintain a bundle of HTML pages in an uniform appearance. Adopt the light and agile approach to document a product or to create a mini-site. Target users are the developers of open source projects.

Everything is very simple, this tool is just a Python wrapper for XSLT, but:

* There is a big difference between “easy to do” and “is done, with documentation”.
* The main feature is the descritipion of use cases. They highlight why 99.99% of other CMS don’t fit to the needs of the open source developers.
* Also useful: patterns for simple templating tasks.

I use the tool for my own projects, and I’m very satisfied.

Start a browser, login to picasaweb. Create the album “incoming”. Go to the photo upload page. View the page source (or use Web Developer plugin to get form details). Follow the data in the HTML code to create picasa.sh. Here is the skeleton:

curl \
-b /home/olpa/.mozilla/..../cookies.txt \
-F redir='http://picasaweb.google.com/..../Incoming?authkey=....' \
-F uname=.... \
-F aid=.... \
-F num='1' \
-F name='' \
-F file0=@$1 \
'http://picasaweb.google.com/lh/addPhotos?tok=....'

Option “-b” is the path to your Firefox cookies file. The value of the form field “file0” is read from a file. The value “num” shouldn’t be empty, but set to “1“.

Usage:

$ picasa.sh my_photo.jpg

If everything is ok, you’ll get as the response “Moved Temporarily”.

To upload all the files in a directory:

$ for i in *jpg; do picasa.sh $i; done

evil_empire++

I’ve realized that there is no such feature as password protection of albums. Picasaweb disqualified, I’m going to search for another web album service.

Download and install yet another plugin, Executable PHP widget. Then create a PHP block for the sidebar, and put as the content something like:

<?php if (function_exists('wpads')) { wpads('sidebar'); } ?>

Don’t forget to save the changes!

The best I found is “Blog Post Category Trauma: How To Help Bloggers With Useless Categories“. Unfortunately, this text explans only that bloggers have to have good categories, but doesn’t explain how to achieve this goal.

Finally, I invented an ultimate method. Inventing categories and tags (labels) is no more a problem.

From Kipling, The Elephant’s Child:

I keep six honest serving-men
(They taught me all I knew);
Their names are What and Why and When
And How and Where and Who.
I send them over land and sea,
I send them east and west;
But after they have worked for me,
I give them all a rest.

I let them rest from nine till five,
For I am busy then,
As well as breakfast, lunch, and tea,
For they are hungry men.
But different folk have different views;
I know a person small-
She keeps ten million serving-men,
Who get no rest at all!

She sends’em abroad on her own affairs,
From the second she opens her eyes-
One million Hows, two million Wheres,
And seven million Whys!

This verse inspired me to try this method:

* Print all the post titles.
* Near each title, write a one-word comments about content of the post, answering the questions “what, why, when, how, where and who”.
* If the same word or its synonym is used for three posts, it’s a category.

And yes, this method worked fine. I easily got 11 categories.

What’s more important is that the method helps to realize categories which can’t be found by usual thinking. For example, I even didn’t suspect I have category “usability”.

Another useful way of generating .dtx files is to write the documentation and the code separately, and then to combine them using the makedtx system. This technique has particular value in that the documentation file can be used separately to generate HTML output; it is often quite difficult to make LaTeX to HTML conversion tools deal with .dtx files, since they use an unusual class file.

* Space: above, below:
\addvspace

* Align: left, right, center, justified
ragged2e package
\Centering \RaggedRight \RaggedLeft

* Font, linespacing
\selectfont and family

* Keep with next or previous paragraph:
\nobreak or some penalty

Some time ago, Lauri Harpf (A Promotion Guide) published a nice article “How to stop spam on your phpBB forum“, in which he pitched the tools.

I announced the article in the bbAntiSpam forum and the blog “Spam Bots and CAPTCHAs“.

In a week, the article appeared on the Google’s first page.