Brandon Phoenix - aka Dim on Blenderartists.org has completed a huge document detailing the many aspects of Video Games. The document is a great reference and learning tool for anyone interested in...
[read the full article on blendernation.com]
Brandon Phoenix - aka Dim on Blenderartists.org has completed a huge document detailing the many aspects of Video Games. The document is a great reference and learning tool for anyone interested in...
[read the full article on blendernation.com]
I acknowledge that nothing here is new, but I’ve tried to go from first principles. IIRC, Ralph Meijer has already had some of this stuff working.
Okay… Instead of users being referred to by a profile URI, which seems a bit sucky to me, let’s start off by assuming there’s only one microblogging site per domain. I think that’s a reasonable restriction.
This allows you to refer to a particular user globally by a tuple consisting of a username and a domain. Are we good so far?
As far as I can tell, existing microblogging sites don’t allow an @ sign in their usernames, so this allows us to have a string representation for any user, globally, of:
user@domain
Wowee, groundbreaking.
So assuming that’s all okay so far, we need a protocol to carry data from one domain to another. Given that we want to avoid polling, we’d be best off with a continuous stream orientated protocol, rather than a request/response one. A TCP based protocol seems sensible, since presumably neither end is going to be behind a heavily restricted NAT, and so communications are going to be reasonably free.
The protocol probably needs to be extensible, which suggests to me that we want to be looking for an XML based protocol. And that rather suggests that XMPP or BEEP is a solution here. BEEP is actually lighter - as long as we strip out the over-engineered stream bits we don’t want - but there are many fewer libraries, and besides, most of these sites have an XMPP interface anyway.
So, this suggests that each microblogging update is carried in a single stanza, and that the “subscriptions” are effectively treated as a roster subscription.
So far, we have XMPP, with some kind of message (we’ve not yet decided what) whizzing between servers.
Next, let’s consider what existing facilities are present in XMPP which might do this job.
The obvious one is PEP. In PEP, each user has a number of nodes, which are individually addressable, and can effective act as broadcast demultiplexers - the user emits a message - an XML gobbet - “aimed at” the node, and the node emits a message to each subscriber.
It’s simple, and almost exactly what we want, so if nobody objects we’ll go with this.
And there we have an almost complete OpenMicroblogging specification.
The missing bits are:
- discovering metadata about the user, which can probably be done with s to the user, or possibly XEP-0154 attributes.
- the precise formatting of the microblogging updates.
- an optimization for reducing server-server traffic loads.
The latter can be done by, in the case where the microblogging update is open and public (ie, has a predictable and largely uncontrolled access model), then we may as well send updates conceptually between servers, such that multiple users on remote servers “share” a single update message.
The easy way to do this is for the remote server to indicate that the subscription is being proxied.
The thing I wonder about, though, is why one would bother have microblogging at all, in this circumstance - what I’ve outlined above is, after all, essentially a simple PEP service that could be put into place today with many servers, and would take minimal client development. That would mean, though, that microblogging sites would essentially act as an interface onto the XMPP service, which is an odd state of affairs, but I think overall I prefer it.
Pointless Keyword Of The Day: Rick Moranis.
I'm in Istanbul since yesterday. The weather is hot.
We have drunk beer on the sidewalk sitting on cushions and carpet: that's what the terraces are in Istanbul. Some North American countries really need some thinking about the alcohol taboo.
I have taken 346 pictures so far. That's a case where I wish I had a GPS to geotag them. I will need to do some sorting though, and that might involve writing code before hand.
More later...
This update comes courtesy of free wireless at Changi Airport, where I'm transiting on my way to Istanbul. The wave of humidity as you come off the aircraft at Singapore is always familiar and inviting. The Indian vegetarian restaurants in the terminals make me so happy. I am having a 10pm snack of samosa. Flying always makes me hungry. Mmm, cardamomy.
It is Stephanie's birthday today (meaning I left the country), so last night we went to the Fly By Night club in Fremantle to see Peter Combe. Peter Combe was a seminal part of so many Australian children's formative years. His revival just goes to demonstrate how we're all really still kids at heart.
There has finally been an official webkit release for the Gtk+ port, and I finally took the necessary time to package it.
As you can guess from the screenshot above, it now has plugins support.
It is also über fast. You can compare with previous results for both gecko and webkit.
IF you want to try another groupware server, I finally made up ebuilds for Scalable OpenGroupware.org, SOGo for short. These are available in the GNUstep overlay for now (it seems to work fine, but I lack the full server installation needed to completely test it, mostly an IMAP server with LDAP backend).
From the SOGO folks:
SOGo is a free and modern scalable groupware server. It offers shared calendars, address books and emails through your favorite Web browser or by using a native client such as Mozilla Thunderbird and Lightning. SOGo is standard-compliant and supports CalDAV, CardDAV, GroupDAV and reuses existing IMAP, SMTP and database servers - making the solution easy to deploy and interoperable with many applications.
For the curious, a demo web site is running here. Feedback appreciated of
course if you try it on Gentoo
On other news, I noticed that the Window Maker web site is back up, stating:
windowmaker.info Back Online posted on 2008-06-30 09:04:16 by kairi
windowmaker.info has been brought online as of early July, 2008. We are currently working on reimplementing the site in a more modern, safe fashion, while at the same time restoring all services required for development and communication. With that said, we are working very hard to revitalize Window Maker's presence on X Window (and perhaps beyond) desktops. With this new focus, we can now truly assert that Window Maker will be resuming active development very soon. We expect to once again provide the de-facto minimalist yet extremely functional window manager to the world.
This has been my main window manager since... well a loooong time, I'm
crossing fingers and hoping it will really come back from the dead projects
world!
Oh, and if you did not get the news, 2008.0 is here! Thanks release engineering team members
I almost forgot: congratulations to the new Council members, both veterans and newcomers!
Seems with the latest versions of webkit-gtk and midori, a long-standing crasher-bug got fixed and it now allows you to run the browser-test ACID3.
I just bumped the webkit-gtk ebuild in Gentoo to the latest snapshot.
ACID3 is a test for the standards compliance of modern web browsers. I wrote about ACID2 some years ago.
The first, experimental release of Kexi Web Forms is finally here. Here are the features implemented so far:
During the following days i’ll switch to a new server backend (pion-network-library) which is a pure C++ network library written using boost and asio. This will allow usage of pure classes as handlers thus simplifying the code-base and reducing code duplication and, best of all, get rid of all those function pointers. We’ll probably have to ship our (stripped) version of pion-network-library which contains fixes for gcc 4.3.0 and patches to bind handlers to regexp-driven URI patterns. Hopefully these changes will make in upstream very soon, so the local copy is needed only for a short time.
Some new, important features are going to be implemented before the end of the summer too! User management, more user interface refinements and new, important features such as full binary blob support (some code to support image loading and displaying is there but it’s not that usable).
But hey! The post is not finished yet, here’s the screencast I promised you some days ago:
(note: if you are using a feed aggregator you won’t probably see this youtube viewport)
And here’s the high quality video, you can “dragonplay” it by issuing
on the command line. Enjoy and….
Stay tuned 
In our April Issue An Almost Idiot's Guide to PostgreSQL YUM we covered using the new PostgreSQL Yum repository to install the PostgreSQL 8.3.1 release on Fedora, RedHat Enterprise, and CentOS. We also received numerous useful feedback from others on issues they ran into and how they overcame them. The blog comments are definitely worth a read.
Now that 8.3.3 has come out, many of you should be considering upgrading if you haven't already since there are a couple of bug fixes as outlined in http://www.postgresql.org/docs/8.3/static/release-8-3-2.html, http://www.postgresql.org/docs/8.3/static/release-8-3-3.html, and for those running 8.3.0 you will need to reindex your tables after as noted in http://www.postgresql.org/docs/8.3/static/release-8-3-1.html. If you are running version 8.3.1 and above then this is a fairly painless upgrade that just requires you to backup your data as a precautionary measure, but doesn't require a dump reload.
Death and taxes are supposedly the two constants in life. I’ve had to file my first Hong Kong tax return this week. This was simplicity in itself. Having gone through similar processes in the UK, and dealing with HMRC, the comparison (if you can call it that) is very stark. The brown envelope with Inland [...]
There seems to be an interesting attraction between Erlang and Lisp and several times it has been tried to marry them, in different ways. Bill Clementson wrote about it already in his article "Concurrent/Parallel Programming - The Next Generation". Here is an updated list:
And this list does not even include projects like (again defunct?) Kali Scheme, which are clearly related.
Personally, I think that the reimplementation approach will have a tough stance against integration approaches like Distel. They lock out either one or the other of the two language eco-systems: libraries, development tools, etc., and recreating this is a lot of work (but don't let that stop you!) With Distel, I can choose to program parts of the application in ELisp or Erlang, whatever is a better fit.
Distel implements Erlang's on-the-wire protocol which is nice because there is no need to mess around with a Foreign Function Interface. Alternatively, one could bind to the Erlang C libraries. As far as rapid prototyping is concerned, this should be the fastest and most straight-forward approach. I wonder why everybody is doing it the hard way?
I said this briefly on the gentoo-dev list, and I want to expand on it here. The council is Gentoo’s leadership, and it’s composed of 7 people elected by all Gentoo developers using a Condorcet-style ranking vote. Of the 7 people on the old council, 5 of them decided to run again and every single council member who ran was re-elected.
This is significant because the re-election was forced over miscommunication about a meeting, and this created some serious conflicts with a sentence in the Gentoo Linux Enhancement Proposal that created the current structure, including the council. I consider this a mandate, showing that Gentoo developers have confidence in the existing leadership doing what’s best for Gentoo.
This was your chance to say yes or no, and you gave us a resounding yes. Since it isn’t often we hear much from the vast majority of developers, this really means a lot to me in saying which directions we should go, based on who you voted for (graphs here). My interpretation is that you like what’s going on now and where we’re talking about going. I’d really love to hear more input from those of you who don’t normally speak up, though. What can we do for you?
P.S. The 2008.0 release is out.
This extension will add a new item to the File - Templates menu that allows to assign a new template to the current document. All styles and formatting will be loaded from that template and the document will behave as it was created using that template.
New menu is File - Templates - Assign new...
As merging menu items is supported for OOo 2.3 and higher only, there will be a toolbar button to assign a new template.
This behaviour has been requested in Issue 52783.
The extension itself is localized to Danish, Dutch, English, German, Italian, Russian, Serbian and Spanish. If there is need for further translations, please send me an email (you should be able to translate about 10 strings to the desired language).
QuickSort. It’s widely accepted as the fastest generic sorting algorithm around. It’s also very simple. A comment in the last article showed me this Haskell beauty:
qsort [] = [] qsort (x:xs) = qsort (filter (< x) xs) ++ [x] ++ qsort (filter (=">=">>= x) xs)
While my C implementations aren’t any where as beautiful, I’m quite satified with their efficiencies. I’ll keep the code to a minimum, and the let the graphs do the talking.
Introduction
The quicksort is kinda like the heapsort, only not as strong (thus not as reliable). It is generally always recursive, and basically consists of 2 steps (Divide and Conquer):
Conquer: Randomly choose an element x in the sorting set X. Seperate the set into two segments consisting of those lesser than and greater than x, X.1 and X.2
X = [4,7,2,6,89,3,5,7,12], x = 7
X.1 = [4,2,6,3,5] X.2 = [7,89,12]
Divide: Do the same for X.1 and X.2
This will terminate when the size of X.1 and X.2 become less than 2.
A reasonable implementation of this can save on space in comparison with the mergesort (O(n)), with O(1) overhead. Interestingly, this algorithm does not guarantee O(nlog(n)) time; it has a worst running time of O(n^2), no better than Insertion Sort. It just happens that the average running time is O(nlog(n)), with much better coefficients than any other sort (actually, this is only true for comparison sorts, i.e. Heap, Merge, Binary, etc.).
A note, all these implementations have been run while running a normal gnome deskop, with just XChat and mocp running in the background. Let the graphs begin.
Naive Vanilla Implementation (using a scratch buffer):
Key Lines
void sort (int len, int *list) {
…
pivot = rand()%len;
…
if (list[i] < list[pivot]) buf[idx1++] = list[i];
else buf[idx2--] = list[i];
…
Interestingly, this is not half as jumpy as merge sort, considering that merge sort was tightly asympototically bound to n*log(n) (i.e. it’ll stay to that value as you go higher and higher), and this implementation uses random numbers to generate the pivot. The performance is much worse than MergeSort (which had a worst performance of about 0.9 ms)
Single Array Implementation:
Key Lines
SWAP (list[pivot], list[len-1]);
…
if (list[i] < list[len -1]) {
SWAP (list[idx], list[i]);
idx++;
}
…
SWAP (list[idx], list[len-1]);
Much better, though still a bit off the 0.9 of MergeSort. The method swaps the pivot element to the end of the list, and keeps track of the position of the last number smaller than the pivot (or actually, the position after it). And at the end of the algorithm, swaps the pivot into that location.
Median Pivoting Implemenation:
Key Lines
pivot = (len >> 1);
Aha, <i>just</i> better than MergeSort. And all that just by replacing the rather heavy rand() function with a simple median (just take the middle element. len>>2 is equivalent to len/2). Interesting to note the optimization difference.
Grouping Equal Elements Implemenation:
Key Lines
pivot = (len >> 1);
Whoa! <i>Huge</i> improvements. This was with grouping the terms that were equal to the pivot, and excluding them from the Divide step. Notably, this makes much less difference on a random sort.
[BEST] Insertion Sort Implemenation:
median/group-equals/insertion/quick.c
Key Lines
/* Insertion Sort */
if (len < cutoff)
{
for (idx = 0; idx< len; idx++)
{
i = idx;
for (tmp = idx; tmp< len; tmp++) i = (list[tmp] < list[i])?tmp:i;
if (i == idx) continue;
tmp = list[idx];
list[idx] = list[i];
list[i] = tmp;
}
return;
}
A minor incremental improvement. It just grazes 0.8 ms. It seems that QuickSort is more amenable to the insertion sort optimization than MergeSort.
I also have a graph for the single array implementation using the insertion sort.
Looks a bit chaotic, but that might be due to some freak processor usage.
Another set of implementations can be found at Jeffery Stedfast’s place. One thing he tried out was to ‘unrecurse’ it by using a stack containing the extents of any stop (essentially, where the sub array you are sorting in one of the Divide steps begins and ends). I don’t see how this is different from what C is anyway doing, i.e. pushing the variables of a function call to a register, and then popping them later. There would be only the function call step (which is a jmp command, something that takes about 1-2 processor operations. On my machine (3.04Ghz) that would be about 3e-10 seconds). On trying to implement it, I found my implemention was probably very sucky, and was much slower than the un-modified program. However, for what it’s worth the previous implmentations (using a median) are all quite a bit faster that Stedfasts implementation (~1.04s on an array of size 5e6). According tho his post, his function is a bit faster than the glib qsort() function, and that also means mine is too :-).
Well, that’s all I have. You can of course find more graphs and the source code for all of these (and some boring modifications) at /quick/
After the release of Firefox 3 and in preparation for Firefox 3.1 / Mozilla 1.9.1, the trunk has switched from CVS to Mercurial for Firefox, Gecko, Toolkit and most other Mozilla core components. The new repository is called mozilla-central. SeaMonkey, Thunderbird, Calendar and other projects haven't made the switch yet but at least for the three mentioned above, a Mercurial repository called comm-central has been be set up but not yet filled. In the meantime, SeaMonkey and MailNews development continues on the CVS trunk (sort of 1.9.0.x). The below is taken from there, ignoring Mercurial for now (Bugzilla integration, anyone?).
Progress
Meanwhile having a rest from various celebrations, I managed
to complete receiving multhreaded crypto processing
in POHMELFS.
So far it was only tested in debug environment (i.e. zillions
of logs and overall miserable performance), but it shows, that
different threads pick up the work, both on sending and receiving
directions.
There is a limitation though: the same crypto threads are used both
for receiving and transmit pathes, so it is possible to saturate them
all for example for receiving, so sending will stall. If there are
unsufficient crypto threads, waiting for RX crypto processing can take
too long, so watchdog transmit scanner will fire up and complete transactions
with errors. One can work this around by specifying big enough number of
crypto threads or long enough transaction scanning timeout, both are provided
via mount option.
I would like to test it in more production-like environment and perform various
stresses on it, but I'm far from my working place, so can not do it right now.
Which means release will be postponed for tomorrow (if testing will not show
regressions or bugs).
This will not be last feature release though: for example POHMELFS does not support
extended attributes and ACLs, there is no header checksum (although there is a reserved
32-but field) there may be some features in different areas too,
but I do not hurry to implement them, since I need something to put into future
POHMELFS changelogs. I think sending the same kernel patch with different words
about userspace server changes is not the way to go, so there should be some kernel
changes too :)
I will draw up some design notes on how I plan to implement POHMELFS server, and namely
how distributed facilities will be done, so far I have quite clear picture in mind,
but it needs to be worked out 'on paper' to find rough corners.
Stay tuned!
Comments (0)
After more than 3/4 of a year of working (and setbacks), it's finally done: Gentoo Linux 2008.0 install media have been released this morning. Enjoy!
* Personal:
- I am almost settled after one month at Ahmedabad and now early looking for new rented house (ie home), 1 or 2BHK.
- Kavin got minor electric shock while playing with switch and socket. Fortunately, my wife was holding him and doctor told that there is nothing to worry. He is `difficult` to handle and long examination of my fatherhood will start soon when they will be here at Ahmedabad.
- Work is going great. I have very good team now!
* Debian:
- Sandro Tosi will be co-maintainer or probably take over sitecopy package from me and will put it under collab-maint repository. Since, I mentioned that I am bit busy with settling down here and it will not be great if such useful package requires time to fix bugs with patches available already.
- D-I translation of Level 1 and 2 for Gujarati language are again 100%. Hunting Level 3 now.
- I am not still able to install Debian properly on my macbook! Only useful source for this is, http://wiki.debian.org/MacBook Any other pointers/docs will be 
The 2008.0 final release is out! Code-named "It's got what plants crave," this release contains numerous new features including an updated installer, improved hardware support, a complete rework of profiles, and a move to Xfce instead of GNOME on the LiveCD. LiveDVDs are not available for x86 or amd64, although they may become available in the future. The 2008.0 release also includes updated versions of many packages already available in your ebuild tree.
A big thanks goes out to our release engineering team members for their hard work over many months to turn 2008.0 into reality.
Get the new release from our "Get Gentoo!" page.
This extension provides syntax highlighting features for OpenOffice.org. COOoder will be usefull for developers wanting to present code fragments in writer documents. The colors used are the same than the ones from GeSHi. Character styles will be created for each language element.
Arrived in Istanbul on Friday in preparation for GUADEC. Had a great time so far visiting the main attractions like Haga Sofia, The Blue Mosque, the Sultans palace and the underground water cistern. Last night Wim, Tim, Edward and myself went out to met Jan and Jaime for some food. A lot of other people ended up there too and it was nice seeing people again. Turns our there is a really nice street close to the Golden Horn Hotel which provide you with a lot of pillows to sit on as you see in the picture below:
I got myself a Lumix TZ4 camera just before leaving the UK as I wanted to be able to take some good pictures. Compared to my earlier cameras this camera is a huge step up. 10x optical zoom makes a world of difference in terms of what kind of pictures I can take. Was also very happy last night to find that GStreamer is able to play the .mov files generated by the camera easily. Some time ago now we did a call for people to provide us with camera video files, and it seems that work has paid of in handling a lot of the semi-standard mov files that cameras create.
Today we have gathered in front of our hotel for a little impromptu GStreamer summit. So far discussion has been about Git and the possibility of migrating to it from our current CVS repository. While using CVS do give us an air of age, wisdom and venerability there is an inkling in the GStreamer community that we might be using a slightly outdated version control system 
Gentoo Linux 2008.0 is out. "Code-named 'It's got what plants crave,' this release contains numerous new features including an updated installer, improved hardware support, a complete rework of profiles, and a move to Xfce instead of GNOME on the LiveCD."
Funny to see that claiming that the following code is readable stirred up quite a few comments.
std::transform( list.begin(), list.end(), std::back_inserter( res ), std::bind2nd( std::plus<int>(), 42 ) );
During my istanbul stay, the best way to reliably reach me in a timely manner will be text message at +16172307951. Please include your name so i know who i am texting with :)
My email access is spotty so far, and voice mail is hard/expensive to check. I am also staying at the golden horn sirkeci if you want to try to ambush me in the lobby. :)
In Heathrow for another hour, then will arrive in Istanbul for GUADEC. I'm staying at Golden Horn. Guys, lets meet at the lobby around 9PM for mild beer tasting.
Also, it's a shame to read "Istambul" on pgo so frequently. Please, write "Istanbul", read as you wish :).
Can't wait to meet everyone... For those of you not coming this year, sorry guys, have to live with it for a week. I'm talking to you sri :-D.
Elections just ended for the Gentoo council for the next year. Turnout was 57% with 145 developers voting, which is quite excellent. The council, created by GLEP 39 to replace Gentoo's previous hierarchy, decides on global issues and policies that affect multiple projects. To select council members, Gentoo uses the Condorcet voting method, which involves ranking them in order rather than just picking a single candidate. Here are your new council members, listed by ranking in the election results:
All of the previous council members who ran again were re-elected, and the two new members are Mark Loeser and Tobias Scherbaum. A full list of ranked candidates is also available. The new council members will get right to work—the new council's first meeting, scheduled for July 10, is approaching fast.
Enabling PHP-GTK2 code completion in Eclipse PDT for a project is very simple. The following example is given on Windows, but the steps should be the same on any supported platform:
A sound caught my ear, tickling my consciousness until I was impelled to locate its source. It was not that I wanted any ice cream from the ice cream truck I was sure was the culprit; it was because it sounded suspiciously like it was playing Come Dancing by the Kinks. When I saw that it was a large garbage truck, I was still more bothered by the calliope-like riffs coming from it than the fact that the garbage truck was playing a song.
Naturally, I whirled around and goosestepped away. After having dinner with what was probably a contracted companion, I was escorted to a massage parlor. While I was being serviced, I stared at the TV while a news program started. Its theme song was a muzak version of Yanni.
Shortly thereafter I made the mistake of getting into a conversation about lavender: the flower, not the color. The musical accompaniment was Elvis Costello. This is not notable.
Then, before I was fondled unpleasantly by an 8-year-old with Down's Syndrome, whose mother was entirely unsympathetic to my plight, I encountered a much smaller garbage truck. It was playing Für Elise. What a contrast! I only want to encounter garbage trucks playing classic rock from now on.
Come dancing; it's only natural, unless you get molested by a mongoloid.
There hasn’t been much of note in terms of big changes or features in edge Rails lately, so this time I’ll leave you to pore over the Rails commit logs for any bug fixes or minor changes that I haven’t pointed out. There has been some work in progress with ActionPack refactoring and multithreading work as well as some activity in ActiveModel too, but nothing really concrete yet (still very much a work in progress).
As usual, be sure to leave any suggestions and criticisms in the comments.
script/server now checks for the availability of Thin and uses it. Pretty convenient if you are using Thin as your production server (and want to run the same when developing). You’ll have to add config.gem 'thin' to your environment.rb first to get this to work.
This patch was contributed by one of the guys at fluxin.
The String#humanize core extension method is used convert strings with underscore, usually table column names, in them to pretty readable text. For example,
"actor_salary".humanize
=> "Actor salary"
"anime_id".humanize
=> "Anime"Sometimes this doesn’t work out so well though, when you have legacy tables or simply “inhumanely” named column names like “act_sal_money” (which is really “Actor salary”, but would be #humanize-d to “Act Sal Money”).
You can now specify custom inflection rules (just like you would for plural/singular/irregular/uncountable inflection rules):
Inflector.inflections do |inflect|
inflect.human /_cnt$/, '\1_count'
inflect.human 'act_sal_money', 'Actor Salary'
endNotice how you can also use a regular expression above to convert columns like “click_cnt” to “Click count”.
Thumbs up to Dan Manges and Pascal Ehlert for this patch.
Pratik has committed a tiny (but really useful) change to ActiveRecord that allows you to specify conditions on a joined table in its own hash. An example would explain it better:
Anime.all(
:joins => :character,
:conditions => {
:active => true,
:characters => { :gender => 'female' }
}
)
The ActiveRecord query above would find all “active” anime with “female” characters.
That’s it for this week’s Living on the Edge – do let me know if you like to see more write-ups on even the minor bug fixes and changes that’d I’d left out this week.
Yesterday I found I was a rusty writer, and managed to write only two pages of questionable quality. I was not pleased with myself. Today, on the other hand, I got into the old rhythm again and wrote 8 nice pages in a few hours. I felt much more comfortable, too. Practice is everything.
Chapter 4 is about 25% done. The book draft reached 120 pages.
In trying to learn more about how to process the human language and how it can be applied in the context of a game, I recently ran across an interesting knowledge base called ConceptNet. ConceptNet is a freely available commonsense knowledgebase and natural-language-processing toolkit which supports many practical textual-reasoning tasks over real-world documents right out-of-the-box It comes
Yey! libtorrent 0.12.2-1 is now uploaded to experimental (thanks bureado for the sponsoring! :D), soon I'll get rtorrent 0.8.2-1 uploaded as well.
I've been using it since a while and has been working great!
In a few hours the buildd network should get them built for all arches, BTW, in this version is included the support for ARM-based machines :) Prepare your NSLU2!
Battelle game engine is an open source project made on Blender 2.45 by Guillaume Coeur and Bertrand Lapraz in order to complete their bachelor degrees at the University of Geneva (CH). This project...
[read the full article on blendernation.com]
Powered by Twitter Tools.
Like a lot of geeks, I've been interested in how the brain works for most of my life. Artificial Intelligence was always one of my interests within computing (and part of what got me interested in linguistics at a very early age).
Within my linguistics research, I've always been interested in models that are biologically plausible so it was a huge delight to read Jeff Hawkins' On Intelligence back in early 2005 and find a theory that was biologically-based and believable from a linguistics point of view. One prominent psycholinguist told me in 2006 that it was one of the most promising theories he'd ever read.
After reading the book, I promptly went out and built a library (as I am wont to do) of about 20 books on general neuroscience, computational neuroscience and the relationship between the brain and language. I started thinking about how to implement the ideas and, after reading some of Jeff's and Dileep George's early papers, augmented the library further with books on Bayesian networks, belief propagation, etc.
When Jeff and Dileep started Numenta and eventually released an early version of their Hierarchical Temporal Memory (HTM) platform in Python, I was particularly excited to try it out, in particular applying it to linguistics. I started the htm-ling mailing list to gather other people interested in applying HTM to models of language. It turned out to be hard to get word out to other people interested in HTM and linguistics, however.
I never got very far with Numenta's code, mostly because there were just too many other things I was working on.
But then a couple of months ago, I found out Numenta was running a workshop / conference. I thought it would be an excellent opportunity for me to (a) get back up to speed with what Numenta was doing and how to use their NuPIC platform; (b) meet other people interested in applying HTM to linguistics.
So a couple of weeks ago, I attended the first Numenta HTM Workshop. I had a great time. It was great to meet Jeff and the rest of the team. Dileep's talk on the algorithms in NuPIC was particularly helpful to me in understanding how things work.
There were a number of people who expressed an interest in the application to linguistics so in the evening I ran a BOF. None of the attendees (as far as I could tell) were linguists by training so I didn't really get to talk too technically from a linguistics perspective. The boost to the mailing list membership hasn't created any more discussion yet either.
But I am still hopeful that an HTM-like approach (whether in the form of NuPIC or some other implementation) might be useful in building biologically-plausible models of language processing.
Too Bad... i would love to go to GUADEC but yeah i'm still a student with spear time but no money. I wish that conference give us some feedback to where to go and where can we improve our Desktop. Maybe next year GUADEC be somewhere near Portugal and i can afford a trip to meet my fellow hackers.
Still maybe some important Open Source Company can give me a job so i can afford a trip to see one of the things i like in my life... OSS.
Yes i need a job...
Article submitted by Caspar Clemens Mierau. Guess what? We still need you to submit good articles about software you like!
It’s time: no reason should prevent you from adding IPv6 connectivity to your machine. Of course it’s still an issue, as most ISPs don’t provide native IPv6. So in most cases the easiest way for you is to set up a tunnel to an IPv6 broker. There are currently several free brokers. I’ll show a simple way of getting IPv6 connectivity with the aiccu and SixXS.
First you have to apply for an account on SixXS. Please note: as a kind of ISP, Sixxs really need valid information from you. You may give them a link to your Xing or LinkedIn profile.
Your application will be checked and (probably) approved. Wait for the mail. After that go to the SixXS website, request a new tunnel, and pick an entry point near you. This step also needs to be approved. Wait for the mail (it takes up to a day).
Now let’s get it running. Install the package aiccu (apt-get install aiccu). During installation you will be asked, which broker you are using. SixXS is already preconfigured, so choose it and input your account information. If everything is fine, aiccu will check SixXS and ask for your tunnel information.
Open a terminal and run ifconfig sixxs—it should show a new network interface with an IPv6 address. Now let’s check IPv6. Open Firefox and go to http://www.kame.net/. If the turtle logo is moving, your are using IPv6, if it does not, you don’t.
You should understand the SixXS credit system. It’s used to limit users in repeating bad actions and to make sure they maintain their tunnels. For example if a static tunnel is down it will cost you some credits, thus you better keep it up. One could see the credit system as a bank, you got a credit limit and you can’t go over it and buy everything you want, but when you earn credits because your tunnel is up you can do a lot with it.
Note that all your IPv6 traffic will be directed through the broker, so you have to take care of the security.
Check http://www.sixxs.net/misc/coolstuff/ for interesting IPv6 content: high traffic news servers, the IPv6 freenode server and so on. Always keep in mind, that not every application is ready for IPv6 and many applications need to be configured for IPv6. With Debian/Ubuntu you should be able to use at least Firefox, Thunderbird, Pan, and Irssi.
aiccu is available in Debian since Etch, and in Ubuntu since Feisty
Happy networking!
