Planet Larry

Ten days ago I complained that there were no good Lisp equivalents of ruby-forum or perlmonks. It looks like someone went and made one. What good timing.

I hope it's a success, and I hope it stays newb-friendly. The amount of fake watch and shoe spam on comp.lang.lisp has reached critical mass.

Speaking of mailing lists, maybe it's just me but I've never found mailing lists to be all that enjoyable to use. They have the benefit of being a sort of lowest common denominator (everyone has email, and you can slap an HTML interface on top of one). They also have the benefit of being distributed to some degree, because everyone who gets the email serves as an archive, and if the main server dies maybe you can recover things. And mailing lists do have less overhead than MBs when it comes to running one, especially a high-traffic one, I would imagine.

But the bad things about mailing lists vs. message boards:

• Message boards are accessible from anywhere that you have a web browser, which is everywhere. Email isn't necessarily accessible from everywhere, unless you use webmail or SSH home and use mutt or something, which not everyone can or wants to do. Or if there's a good web interface on the mailing list.
• You can't do anything more than plaintext, which isn't entirely a bad thing, HTML email is pure evil, but being able to cleanly post images or clickable links or formatted text on a message board is a nice feature.
• Threading never quite works correctly on mailing lists, because eventually someone will hit the wrong button in their mail client and break the thread; whereas on a message board it always works fine.
• You can move threads around between forums on an MB, you can edit threads, you can close threads, you can delete a post if you make a mistake; but mailing lists are write-only, and once you send a message off into the ether it's posted for everyone to see forever, and no one has much control over a list beyond moderating the messages that end up getting through.
• Avatars. Personal profiles. These things make people seem more like people and less like a nameless entity. It's friendlier and more inviting.
• The HTML interfaces people slap on top of mailing list archives are pretty horrible 95% of the time. Probably because most people are using email clients anyways so no one cares. Message boards generally look nice and have nice interfaces for reading and posting.
• Email sucks. Spam filters and bounced messages mean you never quite know if what you just wrote actually made it to the list. Reply to list vs. reply to sender vs. reply to all, etc. are all needless complications. How many times have you seen "UNSUBSCRIBE" sent to everyone on a list? The interface to mailing lists is not intuitive. Whereas you can always see immediately if an MB post worked or not.

And so on. I likes me my message boards.

apparently, there are some issues with linux 2.6 and pearpc.. so i still cant
get the gentoo ppc to start properly. i thought i’d do some studying in the
meantime..

typedef struct {
        volatile unsigned int slock;
} raw_spinlock_t;

static inline void __raw_spin_lock(raw_spinlock_t *lock)
{        asm volatile(\"\n1:\t\"
                     LOCK_PREFIX \" ; decb %0\n\t\"
                     \"jns 3f\n\"
                     \"2:\t\"
                     \"rep;nop\n\t\"
                     \"cmpb $0,%0\n\t\"
                     \"jle 2b\n\t\"
                     \"jmp 1b\n\"
                     \"3:\n\t\"
                     : \"+m\" (lock->slock) : : \"memory\");
}

interpretation of the extended inline asm:

1) : “+m” (locl->slock)
“+m” means that the memory operand ‘lock->slock’ is both input and output.

2) : “memory”
from gcc manual:
If your assembler instructions access memory in an unpredictable
fashion, add `memory’ to the list of clobbered registers. This will
cause GCC to not keep memory values cached in registers across the
assembler instruction and not optimize stores or loads to that memory.

3) LOCK_PREFIX is defined at /usr/src/linux/include/asm/alternative.h:

#define LOCK_PREFIX \
                \".section .smp_locks,\\"a\\"\n\"   \   ; push current section and change section to .smp_locks
                \"  .align 4\n\"                  \   ; align to 4 bytes
                \"  .long 661f\n\" /* address */  \   ; create an entry in the up/smp alternative locking table
                \".previous\n\"                   \   ; pop back the old section
                \"661:\n\tlock; \"                    ; assert processor LOCK# signal

LOCK_PREFIX is supposed to be called within the context of a procedure (executable) which finds its way to a .text section.

i haven’t digged down deep into the internals of ‘alternatives’ yet, but a quick look at kernel/alternative.c :: alternatives_smp_unlock() shows that it’s employing a self modifying code that patches the “lock” instruction with an architecture specific NOP opcode.

without all the extended asm semantics above, the asm roughly translates to:

1:      lock       ; decrement atomically
        decb    %0 ; really decrement now..
        jns     3f ; successful in acquiring lock?
2:      rep        ; nothing
        nop        ; nothing
        cmpb    $0, %0 ; someone must release the lock first.
        jle     2b     ; busy wait...
        jmp     1b     ; lock available! try acquiring lock again.
3:     ; exit __raw_spin_lock()

and then comes the trylock variant

static inline int __raw_spin_trylock(raw_spinlock_t *lock)
{
        char oldval;
        asm volatile(
                \"xchgb %b0,%1\"
                :\"=q\" (oldval), \"+m\" (lock->slock)
                :\"0\" (0) : \"memory\");
        return oldval > 0;
}

interpretation of the extended inline asm:

1) : “=q” (oldval)
output oldval in any a,b,c or d register.

2) : “+m” (locl->slock)
“+m” means that the memory operand ‘lock->slock’ is both input and output.

3) : “0″ (0)
%0 matches the output register for oldval. initialize that with the value 0.

4) : “memory”
from gcc manual:
If your assembler instructions access memory in an unpredictable
fashion, add `memory’ to the list of clobbered registers. This will
cause GCC to not keep memory values cached in registers across the
assembler instruction and not optimize stores or loads to that memory.

__raw_spin_trylock() basically exchanges the current value in raw_spinlock_t->slock with zero and returns true if locking was successful and false if acquiring of lock didn’t succeed.

lastly, we have the raw unlock procedure:

static inline void __raw_spin_unlock(raw_spinlock_t *lock)
{
        char oldval = 1;
	
        asm volatile(\"xchgb %b0, %1\"
                     : \"=q\" (oldval), \"+m\" (lock->slock)
                     : \"0\" (oldval) : \"memory\");
}

i haven’t been able to understand the significance of the “%b” in the assembler template but looking at the asm listing of the template with and without the %b yielded the same intructions, so could anyone shed some light on this one?

basically looks like __raw_spin_trylock(), but the difference is that oldval is now initialized to 1 instead of 0.

there are many variants stemming from the idea of this implementation. raw reader/writer locks which allows multiple readers but only one writer (with no concurrent readers). and then there’s the spinlock layer above the raw spinlock implementation.. but those are for another post.

’til then.

Coding at its most fun is exploratory. It’s exciting to try your hand at something new and see how it develops, choosing a route as you go along. Some poeple like to call this “expanding your ignorance”, to convey that you cannot decide on things you don’t know about, so first you have to become aware - and ignorant - of them. Then you can tackle them. If you want a buzzword for this I suppose you could call this “impulse driven development”.

spiderfetch was driven completely by impulse. The original idea was to get rid of awkward, one-time grep/sed/awk parsing to extract urls from web pages. Then came the impulse “hey, it took so much work to get this working well, why not make it recursive at little added effort”. And from there on countless more impulses happened, to the point that it would be a challenge to recreate the thought process from there to here.

Eventually it landed on a 400 line ruby script that worked quite nicely, supported recipes to drive the spider and various other gimmicks. Because the process was completely driven by impulse, the code became increasingly dense and monolithic as more impulses were realized. And it got to the point where the code worked, but was pretty much a dead end from a development point of view. Generally speaking, the deeper you go into a project, gradually the lesser the ideas have to be to be realized without major changes.

Introducing the web

The most disruptive new impulse was that since we’re spidering anyway, it might be fun to collect these urls in a graph and be able to do little queries on them. At the very least things like “what page did I find this url on” and “how did I get here from the root url” could be useful.

spiderfetch introduces the web, a local representation of the urls the spider has seen, either visited (spidered) or matched by any of the rules. Webs are stored, quite simply, in .web files. Technically speaking, the web is a graph of url nodes, with a hash table frontend for quick lookup and duplicate detection. Every node carries information about incoming urls (locations where this url was found) and outgoing urls (links to other documents), so the path from the root to any given url can be traced.

Detecting file types

Aside from the web impulse, the single biggest flaw in spiderfetch was the lack of logic to deal with filetypes. Filetypes on the web work pretty much as well as they do on your local computer, which means if you rename a .jpg to a .gif, suddenly it’s not a .jpg anymore. File extensions are a very weak form of metadata and largely useless. Just the same with spidering, if you find a url on a page you have no idea what it is. If it ends in .html then it’s probably that, but it can also not have an extension at all. Or it can be misleading, which when taken to perverse lengths (eg. scripts like gallery), does away with .jpgs altogether and encodes everything as .php.

In other words, file extensions tell you nothing that you can actually trust. And that’s a crucial distinction: what information do I have vs what can I trust. In Linux we deal with this using magic. The file command opens the file, reads a portion of it, and scans for well known content that would identify the file as a known type.

For a spider this is a big roadblock, because if you don’t know what urls are actual html files that you want to spider, you have to pretty much download everything. Including potentially large files like videos that are a complete waste of time (and bandwidth). So spiderfetch brings the “magic” principle to spidering. We start a download and wait until we have enough of the file to check the type. If it’s the wrong type, we abort. Right now we only detect html, but there is a potential for extending this with all the information the file command has (this would involve writing a parser for “magic” files, though).

A brand new fetcher

To make filetype detection work, we have to be able to do more than just start a download and wait until it’s done. spiderfetch has a completely new fetcher in pure python (no more calling wget). The fetcher is actually the whole reason why the switch to python happened in the first place. I was looking through the ruby documentation in terms of what I needed from the library and I soon realized it wasn’t cutting it. The http stuff was just too puny. I looked up the same topic in the python docs and immediately realized that it will support what I want to do. In retrospect, the python urllib/httplib library has covered me very well.

The fetcher has to do a lot of error handling on all the various conditions that can occur, which means it also has a much deeper awareness of the possible errors. It’s very useful to know whether a fetch failed on 404 or a dns error. The python library also makes it easy to customize what happens on the various http status codes.

A modular approach

The present python code is a far cry from the abandoned ruby codebase. For starters, it’s three times larger. Python may be a little more verbose than ruby, but the increase is due to a new modularity and most of all, new features. While the ruby code had eventually evolved into one big chunk of code, the python codebase is a number of modules, each of which can be extended quite easily. The spider and fetcher can both be used on their own, there is the new web module to deal with webs, and there is spiderfetch itself. dumpstream has also been rewritten from shellscript to python and has become more reliable.

Grab it from github:

spiderfetch-0.4.0

One of the great things about Qt is that you can compile the code under esoteric OSs (mmh?) like Windows or MacOS. Although I’ve almost never used windows, I did the packaging of yzis. I used the Nullsoft Installer, which is ok.

I had several issues with static linking, and thought maybe some of you could give some help about it. Don’t ask why I want static link, this is not the issue here.

Problem #1:

The most important one is Qt itself. I expected that to be easy, but it is not. The FindQt4 shipped with cmake 2.4.8 or 2.6 is not really aware of static lib, or is it ? Nor is the one from kde svn. I’ve tried using the one from quassel-irc, but it fails as well. If you can help me with this, the code is available.

It seems (according to #qt on irc) that it makes sense to have two different Qt installed, one static and one shared. I did that (double the space used…) and in one of them did “configure -static” and recompiled qt. Now the *.a are much bigger, and though qyzis links against those *.a, it still depends on the DLLs, according to “dependencies walker”. Gr…

Problem #2:

Then, there is the problem of gnuwin32 tools. In the lib/ directory, you can find for example

• libintl.dll.a
• libintl.def
• libintl.lib

And the DLL (libintl3.dll) is in the bin/ directory. I need to ship the DLL with qyzis, because cmake found the dll, and linked against libintl.dll.a. In a perfect world, I would like cmake to use the libintl.lib and have a static link, but i dont know how to convince cmake of doing that. Do you know ?

Problem #3:

How do you find out what kind of library a file is ? According to what I’ve found googling the web, libraries can have a wild number of different names (libX.dll, libX.dll.a, libX.lib, libX.a, all of this without the lib prefix and so on…). Even worse : libX.a for example, could be either a static lib or the stub for the DLL. Do you have clarifications about this ? Do you know of a reliable way to know which kind of library it is ? (DLL, DLL stub, or purely static).

Ok, here it is, things have finally settled and we are happy to bring you the first alpha release for Yzis. The foundations are stabilizing, and we are now focusing on fixing bugs and porting to different architectures, OSs, and interfaces.

What you have so far :

• nyzis (curses frontend) working under linux.
• qyzis (Qt frontend) working under linux and windows.
• A windows installer.
• All tests (both unit tests and lua/script tests) pass.
• Most important bugs in the bugtracker fixed.

The most noteworthy missing feature is the KDE embeddable component, we are aware of this, and this is the main thing we shall work on after 1.0 is out.

At this point, we are not yet asking for broad testing, but developers not afraid of bleeding edge are encouraged to test it out. If you can help with porting/solving bugs, this is of course even better.

You can join us on #yzis on freenode to discuss all of this.

The web site has been updated, and you should find information about getting the source from source control (mercurial) and compiling on most platforms.

http://www.yzis.org

The tarballs and the windows installer can be downloaded from this url:

http://labs.freehackers.org/projects/list_files/yzis

Below are some screenshots showing the different interfaces (qt/linux, ncurses, qt/windows):

I’ve been pushing some new photography to my photoblog over at Sumatro.com.  I have a new set of work lined up, all being published at 9am every morning.  To keep up to date, subscribe to the RSS feed.

Would love some comments/idea/abuse on the photos if you have the time.

Python

PHP

Ruby On Rails

Django

Are you ‘up to date’ if you are still using a static webpage today? I wouldn’t say that there is a global answer to that question and everybody’s got to find out for himself.

Usually I’m a big fan of automatism and technology and its advantages - but is it really worth it? Let’s have a closer look at the topic:

A Content Management System (CMS) is a program, running on your webserver, generating the html output for the visitors browser, which is actually a good thing, if things are changing a lot.

But a CMS is just a piece of software that needs updates and patches too, disregarding new features and stuff as there might be undiscovered security holes which might turn out to be a rather serious problem. I don’t want to question any cms here. I just want to make you think if you really need one.

I’ll give you an example: My domain stargazer.at hosts a portal page, offering links to my services, my blog, whatever. As my services don’t change that often, there is no need for dynamical created content as things just don’t change. If I think back, the last update on the portal page happened four month ago. Since then it wasn’t touched anymore. Regarding security, this would be quite fatal. In the other hand, my blog here gets much more love and changes much more often. That’s why I am using Wordpress here and plain html on my portal page.

Nick was home from college this past week, and I took a lot of time away from working on the blog and BIOSLEVEL to spend some time with friends. It was a good time for sure, but I always come to regret it when I realize how little work I've done.

Hopefully I get things back on track over the next few days. I'm way behind on several reviews for BIOSLEVEL, and some of the products were just released in the last two weeks or so.

I also rebuilt two of my computers in to new housing, with a few new components that were sent to me for review. Let me say that I hate reviewing some CPU coolers. A certain company that will go unnamed for now sent two CPU coolers that otherwise seem impossible to install on an Intel machine. Stay tuned for a review, if I ever get them installed.

On top of all that, I somehow started playing World of Warcraft again.

There's an article on Slashdot about a US Senate hearing on laptop seizures at border crossings. This affects me, because I travel to Canada a lot and plan to move there within a year or so.

It's a problem because my job requires me to handle what amount to people's medical records as data files on my laptop. It's part of my job, and often I work from home. As of right now, I never take my laptop with me to Canada partly because I don't know what would happen if a border agent decided to inspect or copy all of my data. I can get in very serious trouble for breaching patient confidentiality. On the other hand I could get in serious trouble if I refused to allow a search for myself; at best I'd be turned way at the border, having wasted hundreds of dollars to travel there.

I really don't know what I'm going to do when I move. I'll probably have to wipe my computers clean before shipping them up there. Another option would be to encrypt all the data, upload it to the server that hosts my website, then download it all again after I move. It's insane that I'd have to do such a thing though. And shuffling sensitive data around to strangers' computers and servers isn't the safest thing in the world either.

How do lawyers and doctors and people with trade secrets and other people with classified or legally protected information handle border crossings? It's a bit of a conflict of interest.

Ok, I hinted that I would do a more thorough review of GitHub, the new and easy-to-use git repository hosting site. Although I’m still no power-user I’ve come to know the features that make GitHub worth using and so far unique.
The “Fork” feature is probably the most important one. Instead of just cloning a repository and working on it locally, you can fork it on GitHub. When you fork a project everyone can see you did, and has a nice flashy graph-view so they can see where you branched and what commits you made that are not (yet) in the upstream. And if you have introduced changes that you think would benefit the project you can send people (e.g. the original project owner) a “pull-request”. The recipient can then easily fetch/merge your changes into his project. It really doesn’t get much easier to contribute to (open-source) projects. I certainly did for the first time ;)

Then there are feeds. You can watch projects, which means that your feed includes any commits/comments on the watched projects. It goes without saying that this can be quickly overwhelming for active projects. Comments can be made on specific lines in a commit (or on the whole commit), which is a great feature (think of it as the equivalent to patches being discussed line-by-line in mailing-lists). I still prefer mails though ;)
The syntax-highlighting looks pretty good. I’ve already mentioned the very pale interface (as in low-contrast) and it still has not changed. But I think that most people really browse the commit history in their local clones anyway.
Each project has an attached Wiki too, so you can add a few pages (or a lot if you wish).
To conclude I can say that GitHub is a great service, since it has a free plan for public projects. I would not need it for my personal projects, but to run a or contribute to a open-source project it’s perfect. The amount you use its features is up to you. If you already have a Wiki, already have an active mailing-list and discuss patches there too then you can just use GitHub as the central source code repository (plus the forks of course). The real work is done on your local repository with git anyway, but y’already knew that I guess ;)

So the other day I was thinking what a pain it is to handle text in input boxes on web pages, especially when you’re writing something longer. Since I started using vim for coding I’ve become aware of how much more efficient it is to edit when you have keyboard shortcuts to accelerate common input operations.

I discovered a while back that bash has input modes for both vi and emacs and ever since then editing earlier commands is so much easier. And not only does it work in bash, but just as well in anything else, like ipython, irb, whatever.

So now only Firefox remains of my most used applications that still has the problem of stoneage editing, and I’m stuck using the mouse way too much. It bugs me that I can’t do Ctrl+w to kill a word. Thus I went hunting for an emacs extensions and what do you know, of course there is one: Firemacs. Turns out it works well, and it also has keyboard shortcuts for navigation. > gets you to the bottom of the page, no more having to hold down <space>.

Nvidia reiterated that it won’t provide open source drivers for Linux because the company claims there is no need for it.

I had to install amsn just so could video chat with a friend on MSN. Really? On the plus side, at least I could do it :)

Also, figured out how to use skype on Ubuntu. Skype really want /dev/dsp. And Ubuntu now uses pulseaudio. So basically if you've done anything with sound, Skype won't be able to get the sound. Which sucks. However, pulse audio ships with this handy utility 'pasuspender' which temporarily suspends pulse audio and it's lock on /dev/dsp. So to actually use skype

pasuspender skype

And skype can then seize and monopolize the sound card. So you can use it, but no other sound till you shut Skype down. So it's 2008 and we still can't share the sound card :/.

I just found a couple cool programs called turbotail and multitail while searching for rbot using eix.

Turbotail is just like tail, but it uses dnotify instead of auto refreshing a defined number of seconds.  I always thought it was kind of silly to keep refreshing the screen searching for new content with tail.  Turbotail just sits there until the kernel notifies of a change in the file that you are tail’ing and then it updates what you see.

Multitail looks like a VERY robust way of viewing multiple files.  It can tail any number of files and supports text filtering and even syntax highlighting.

Turbotail works great, but unfortunately multitail crashes when I try to run it from my Yakuake console.  I get this:

--*- multitail 5.2.0 (C) 2003-2007 by folkert@vanheusden.com -*--
 
A problem occured at line 511 in function mynewwin (from file term.c):
 
Failed to create window with dimensions 55x9 at offset -27,-4 (terminal size: 167,19)

Seems to work just fine from a regular console though.  It will take me a while to actually learn all of the features of multitail.

After a while of looking for a new job, I finally got a new job. Well, actually, I’ve been working my new job for about 3 months now. So… I guess it’s not really a new job anymore.

I’m now a full time Linux administrator with ExLibris. Unfortunately, Red Hat is the preferred distribution. That’s to be expected. Most business want to make sure they use something that’s proven and has a clear line of support.

The new job is in Des Plaines, Il, which means I’ll have to sell my house and move a little closer. So far, the job seems pretty good. It’s doing something I like and the people are nice, and most of them seem pretty smart.

Inevitably, as you’re learning a new skill, such as a programming language, you may want to revisit your old work and see if you can do it better. I had this pair of functions to fetch the previous and the next “page” of a source collection, based on the sort order of the source. In plain text, in order to get to the previous “page” I want the source with the maximum sort order smaller than the present one, and with the same parent id. Here is my brute-force approach from a couple of years ago:

CREATE OR REPLACE FUNCTION get_prev_page(INTEGER) RETURNS INTEGER AS $$
DECLARE
    self_page INTEGER;
    prev_page INTEGER;
    prev_src INTEGER;
    par_id INTEGER;
BEGIN
    SELECT parent_id FROM sources INTO par_id WHERE source_id = $1;
    SELECT sort_order FROM sources INTO self_page
        WHERE source_id = $1;
    SELECT MAX(sort_order) FROM sources INTO prev_page
        WHERE parent_id = par_id AND sort_order < self_page;
    SELECT source_id FROM sources INTO prev_src
        WHERE parent_id = par_id AND sort_order = prev_page;
    RETURN COALESCE(prev_src,0);
END;
$$ LANGUAGE plpgsql STABLE;

Ugly, ugly, ugly. I remember what was the main stumbling block here: You can’t use an aggregate function such as MAX() in a WHERE clause. The thing is that you don’t need all those assignments. A little code folding, replacing variables with sub-selects, takes you a long way:

CREATE OR REPLACE FUNCTION get_prev_page(INTEGER) RETURNS INTEGER AS $$
DECLARE
    pp INTEGER;
BEGIN
    SELECT source_id INTO pp FROM sources
        WHERE parent_id = (SELECT parent_id FROM sources WHERE source_id = $1)
        AND sort_order < (SELECT sort_order FROM sources WHERE source_id = $1)
        ORDER BY sort_order DESC LIMIT 1;
    RETURN COALESCE(pp, 0);
END
$$ LANGUAGE plpgsql STABLE;

The “ORDER BY sort_order DESC LIMIT 1″ is a great idiom whenever you need to use an extreme value as part of a WHERE clause.

Even if the two versions of the code above are functionally equivalent, and the PostgreSQL planner probably will rewrite the query to something like version 1, most programmers would prefer version 2. Why? I think there’s a lot to the concept that “coding is poetry”. Version 2 is more esthetically pleasing, because it conveys its inner meaning in a much more succinct way, as in the Merriam-Webster definition of the word succinct: “marked by compact precise expression without wasted words”. At least to me it does. And that’s probably some of the essence of poetry.

My server, mindstab.net, runs Hardened Gentoo. I like it. It provides nice features from grsecurity and PaX like memory randomization, non executable writable memory, etc. However, it really doesn't get along so well with Lisp. Lisp in general seems to like executable and writable memory, and SBCL at least also doesn't like randomized memory. So it took a bit of work to get Lisp onto my server.

Approach 1: Failure
I spent a bunch of time trying to patch the build process in portage to coax SBCL into building. First, of course, I used gcc-config to disable the hardened gcc profile, and just use the vanilla one. Then I created a suid root shell script to call "paxctl -m -p -r -e $1" so that the sandboxed build process could disable PaX features on the SBCL binaries. I added the command to the ebuild, and created a patch to insert the command into SBCL's build process. The process goes like this, portage download's the SBCL source and a pre-compiled SBCL binary. The patched ebuild then calls my suid root script which disables PaX on the pre-compiled binary so it actually runs (as opposed to crashing under PaX) and then a new SBCL binary is built from the source and the pre-compiled binary builds a core file from the SBCL lisp source. The patched SBCL make.sh then again calls the suid root script on the new binary, so it will run. Then it should load the new core and recompile the system for itself. Sadly, while it runs at least, it chokes on the core file and hangs while using 100% cpu. I couldn't get past this so I eventually gave up. If anyone has any suggestions that'd be great.

Approach 2: Success
So the actually solution was as follows: Download the most recent precompiled SBCL binary from the website (1.0.15 for x86), run "paxctl -p -e -m -r -x -s " on src/runtime/sbcl (to cover all the bases). Then run "sh install.sh" to install SBCL to /usr/local. That's it.

The problem with this is you can't emerge lisp packages in portage, you have to install them by hand (unless maybe you want to fake inject the package into the portage database).

I downloaded a copy of slime, untarred it and popped it in my .emacs and I had a full lisp environment ready to go, and on my hardened machine no less. Not so bad.

Bus 006 Device 007: ID 08ff:1600 AuthenTec, Inc.

# layman -a wschlich-testing
# echo =app-misc/fprint_demo-0.4 >> /etc/portage/package.keywords/general
# echo =media-libs/libfprint-0.0.6 >> /etc/portage/package.keywords/general
# echo =sys-auth/pam_fprint-0.2 >> /etc/portage/package.keywords/general
# emerge -av fprint_demo pam_fprint

$ fprint_demo

auth sufficient pam_fprint.so

auth sufficient pam_unix.so try_first_pass likeauth nullok

revolver ~ # lspci
00:00.0 Host bridge: Intel Corporation Mobile Memory Controller Hub (rev 03)
00:02.0 VGA compatible controller: Intel Corporation Mobile Integrated Graphics Controller (rev 03)
00:02.1 Display controller: Intel Corporation Mobile Integrated Graphics Controller (rev 03)
00:1a.0 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Contoller #4 (rev 03)
00:1a.1 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #5 (rev 03)
00:1a.7 USB Controller: Intel Corporation 82801H (ICH8 Family) USB2 EHCI Controller #2 (rev 03)
00:1b.0 Audio device: Intel Corporation 82801H (ICH8 Family) HD Audio Controller (rev 03)
00:1c.0 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 1 (rev 03)
00:1c.1 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 2 (rev 03)
00:1c.2 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 3 (rev 03)
00:1c.3 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 4 (rev 03)
00:1d.0 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #1 (rev 03)
00:1d.1 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #2 (rev 03)
00:1d.2 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #3 (rev 03)
00:1d.7 USB Controller: Intel Corporation 82801H (ICH8 Family) USB2 EHCI Controller #1 (rev 03)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev f3)
00:1f.0 ISA bridge: Intel Corporation Mobile LPC Interface Controller (rev 03)
00:1f.1 IDE interface: Intel Corporation Mobile IDE Controller (rev 03)
00:1f.2 SATA controller: Intel Corporation Mobile SATA AHCI Controller (rev 03)
01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 01)
02:00.0 Network controller: Intel Corporation Unknown device 4229 (rev 61)
05:00.0 SATA controller: JMicron Technologies, Inc. JMicron 20360/20363 AHCI Controller (rev 02)

• The fingerprint reader (Bus 006 Device 002: ID 08ff:1600 AuthenTec, Inc.) [EDIT now it works]
  
• The LCD backlight controls (CONFIG_ASUS_LAPTOP=Y doesn't help and acpi4asus doesn't compile)

• ALSA

00:1b.0 Audio device: Intel Corporation 82801H (ICH8 Family) HD Audio Controller (rev 03)
I needed to add "options snd-hda-intel model=lenovo" to /etc/modules.d/alsa and update modules.

• Integrated 1.3 megapixels webcam (I don't really need a webcam)
  
• [edit] now it works, see here
  

• Environmental damage: You produce many CDs which are a pain in the ass to recycle (and no they do not belong into the normal trash or the German "gelber Sack").


• People really interested in the software packages will go to the websites anyways looking for plugins, themes or just updated versions (with the speed in which many software projects move as soon as you download a version and have it burned to CDs there will be a new version out)


• I have yet another coaster around.

I’m releasing part of the code developed during my PhD. Those are tools useful in the field of convex processing, and more precisely to solve a problem called convex feasbility problem.

This is based on Qt4, and released mostly under the GPL (with an added restriction : you must cite my work if you use this). Oh.. and there are unit tests, examples, and documentation, too.

caution : rest of the post is for those mathematically oriented people, and no more related to free software.

The convex feasbility problem is to find a point in the intersection of closed convex sets inside a Hilbert space. Believe it or not, quite some real-life problems can be formulated as such, for examples some image reconstruction problems in medical imaging, or also in the field of signal processing.

There are a lot of algorithms to solve this, you can find more information about those in the bibliography of my papers, or those from my PhD advisor M. Combettes.

This code provides the necessary tools to play and test those algorithms. Besides all elementary methods on Euclidian/Hilbert Space, it provides a way to compute projections on different interesting convex sets. There are also the needed tools to experiment with tomography, which is one of the main application in my PhD. You can compute the radon transform of an image and use this as data in convex feasibility formulation.

This is of course mostly of interest for those working in convex optimization and/or tomography. I want to highligh the fact that sharing such kind of code is very rare in this community. I would have gained a lot of time at the beginning of my thesis if other people had provided such code.

link : http://labs.freehackers.org/wiki/cvx-processing

What's the Common Lisp version of Perlmonks or Ruby-forum? I have yet to find it.

comp.lang.lisp is largely crap. 50% of the traffic on that list is spam about shoes and fake watches. The other half is equally split between:

• People debating tiny, silly semantic points of the Common Lisp Hyperspec.
• People stuck in the 70's or 80's, talking about the good old days, ruminating about Lisp history.
• Flame wars.
• New people asking for help. Some get good honest advice and helpful answers, many are flamed and ridiculed into next week if they even hint that they dislike the parentheses.

The Common Lisp community (if you can call it that) is a bunch of really smart guys, but they all live isolated in hermit shacks up in the mountains and they spend their time doing magic tricks with Lisp that few people ever see, and if you wander too close they throw rocks at you.

What's the Common Lisp equivalent of perldoc or rdoc? We have the Hyperspec. It's an impressive document, but it's a bunch of painful HTML that looks like it was created in the early 90's, probably because it was. It reads like a dusty, dry, technical document probably because it is. What it's not, is friendly or easily readable.

Perl has CPAN, Ruby has rubygems, what does Lisp have? Either a hand-rolled system definition script, or if you're lucky an ASDF install file. ASDF is the semi-standard Lisp way of installing libraries, except that it doesn't quite work in Windows, it doesn't check dependencies or handle different versions of a package very well, and it doesn't work the same on all Lisp implementations. Many people in the so-called community think it's not very good.

The fellow running Lispcast makes another good point. Where can you download Lisp? It's not obvious.

You could say "OK Brian, good idea, now get to work!" The problem is that even if I had the time or willpower, I'm not the smartest guy in the world. I honestly don't think I could design and run and maintain a CPAN. And even if I did, would anyone use it? But I do know that there ARE plenty of smart, enthusiastic people using Lisp. Yet high-quality friendly code is largely not being produced.

Peter Christensen wrote about "langauge snobs" and the importance of community. One point made is that some really ugly, horrific languages have been extremely successful simply because they've been accessible and fun. An example given is the scripting language in Second Life, which has over 2.5 billion lines of code written in by tens of thousands of amateurs and has accurately modeled a realistic 3D environment with thousands of users at any given time. All in an ugly language some guy invented AND implemented in one week. The developers admit that the language is total crap, but it doesn't matter. 1) It has very good and accessible documentation, 2) it has a very newbie-friendly community, and 3) and it's easy to pick up, throw together some code and get immediate results. Three things Common Lisp lacks.

This is something I've said myself many times: an active, supportive, enthusiastic community is essential for the health of any programming language. Common Lisp simply doesn't have one and it's a shame.

I still secretly hope that Clojure or NewLisp or Arc turn out to be a huge success. They are the kinds of things Lisp needs today.

Sun admitted that Solaris’ late move to open source is a big mistake. They said that while they were busy making money, they were already losing the server market, Linux is taking over very fast.

Having tried several versions of Solaris and OpenSolaris, in my honest opinion, opening its source code to the public is a good move.

What’s cool in OpenSolaris 2008.5?

1. IPS - a network package management system that resembles an apt-get type of command. Thanks to Ian Murdock.

2. Bourne Again Shell - C Shell is no longer the default one. This makes majority of the Linux users comfortable when shifting to OpenSolaris.

3. OpenSolaris Developer Expert Assistance - A dedicated online support service for developers that provides technical assistance for code support, programming questions, diagnostic advice, how-to’s and best practice guidance.

4. OpenSolaris Subscription Support - Telephone and online technical support. Provides automatic notification of security updates.

They asked for it - they got it… a nice DDoS (spreadfirefox.com and mozilla.com were unreachable for a couple of minutes after 17:00 UTC).

Anyway… go ahead and download it, it’s a good piece of software. I tried out beta3 and the RCs, and I like the new version so much, that I’m now using Firefox as my default browser again (after more than two years with Konqueror as my default browser).

Nice birthday present by the way, Mozilla! Thanks!

The best browser ever, landed on earth today but while big daddy was busy asking everyone to do their homework to get ready for the party, he failed to do his own.

First and foremost, as all of us came know a little later, the download day was not just 17th June, 2008; it was 17th June 2008 at 10:00 am PDT. No one ever cared to inform that important piece of shit. All big daddy wanted was a 1 million downloads in 24 hours, he didn't care how frustrated you would be to find out that you don't see your download at 12 am of June 17th, local time. And no Sir, I am not asking you to take care of all the time zones here, a simple countdown timer on the website would have been enough. Moral of the story: Big Daddy, you came up with something really impressive, you made us do our homework, but you failed to do your own!

Now, everyone knows that the digg effect is not just limited to sites on Digg's front page. Anyone can feel it; especially when you want 1 million downloaders to come to your site when you release the most anticipated browser ever, what on earth were you thinking ?! It's never going to be a breeze. Everyone knows that! Again, you failed to do your homework! Moral of the story: You promised us all the good things on June 17th, and all the good things you did give, but; the overall experience was somewhat dissapointing :(

Ok, now that my rant is over, I am off to start enjoying my brand new browser :)

read more

I purchased my MacBook back in November of 2007, shortly after the Santa Rosa chipset update. I'd use Mac OS X a little previously, mostly in the form of trying out OSX86 on my old laptop. After using it for several months as my primary mobile platform, I decided it was time for something different.

OS X is nice and all. User-friendly, intuitive, and works fantastic with the hardware Apple ships. I still feel it's lacking in some areas where Linux excels, such as configurability and the space for customization. Not to mention that there's essentially no decent games out for OS X. I can install Linux and throw on a native version of UT2004 and play away now.

Installation wasn't horribly difficult, but Mac OS X was needed at least initially, and I didn't realize this. Gentoo now boots without an issue on the MacBook. All that I need to finish doing is installing the programs I need and setting up the WiFi.

I'm hoping that I have all the hardware working correctly in the next week or so, and that includes the web cam and touchpad. My only other concern is battery life. It'd be fantastic if I get the same amount or more battery out of the machine with Linux.

Friday a guy on the phone said he'd call me back Monday or Tuesday to give me an update on when / whether they're ever going to send me my monitor. Monday came and went with no call. Not really surprising.

I filed a complaint with the BBB today. We'll see how that goes. At the BBB Westinghouse has around 150 complaints in the past 36 months, but 133 of them were supposedly solved "satisfactorily" and Westinghouse somehow still has the highest possible rating at the BBB. I've read some things about the BBB not being an entirely neutral entity itself, but who knows. I'll start filing complaints with other consumer groups if I need to.

A good handful of people have left comments here at my blog saying they aren't going to buy anything from Westinghouse themselves, which is great to hear. I may mention my blog to Westinghouse next time I call them, if there is a next time. Is not sending me the monitor I paid for really worth losing a bunch of customers?

The sad thing is that I really do need a monitor with component and composite inputs, and they are somewhat rare (the local store had none except Westingcrap brand). However I have found a Gateway model that has them, so maybe that'll work out. I'd gladly take a refund from Westinghouse rather than a monitor at this point.