Search Results: "ag"

15 January 2007

Biella Coleman: Absence makes the heart grow fonder

So so so so I am in the airport in Minneapolis, about 2/3 into my mega-trip from PR to Edmonton and decided to jump online since I have over 3 hrs in the airport. And today is the first day in about a month I have used high speed Internet access and it is, well, beautiful, at least compared to the dial-up I was using at home, which was pretty painful, especially because the slowness was compounded by the mosquitoes flying above me, reminding me of how annoying it was to be online. So I usually ditched the whole experience in favor of the more reliable meat world. Here and nw the speed almost feels magnified. It is like faster, better.. plainly hotter. It reminds me of the time that I quit coffee for months and then I started again. It was like drinking coffee and caffeine for the first time all over again. So I gues what I should try next is to quit coffee and high speed internet together for a few months and then start them together. :-)

Wouter Verhelst: FOSDEM key signing party

I was appointed by the FOSDEM orga team to be the "volunteer" to organize the FOSDEM key signing party again this year, because "I did it so well last time". Hah. In other news, I still have keys to sign from the DebConf5 and FOSDEM 2006 key signing parties. As a result, I had resolved not to participate in any key signings anymore until I caught up my backlog. Yes, I know about caff, and yes, I'm using it, but I still have a lot to do. So now I'm in the strange position of having to organize a key signing party that I'm not going to attend (at least not until the end). Or perhaps I might do so after all... otherwise it's going to be very boring. I'll see. For those who care, the announcement has been out since yesterday (Sunday) and can be read on the FOSDEM website, complete with instructions. In the day and a half that this was true at the time of this writing, 15 people already submitted keys. That's quick.

Steve Kemp: Would you choose to walk or stay

Once upon a time I wrote a collection of shell scripts for managing virtual hosting. Combined these allow me to “easily”: These work well, but every now and again I think “I want a web-based control panel”. After mulling this over for a while and remembering how many of the existing control panels are tricky I’m actually going to do it. There are two hard parts: The last point is the part I’m worrying about right now. So far I have a list of mail domains, and a list of web domains - and the two are unrelated. This becomes obvious when they are listed:
Domain Facilities
steve.org.uk MAIL
bar.steve.org.uk WEB
foo.steve.org.uk WEB
See? No link - I’d expect to see the domains listed as just “steve.org.uk” with links to “Mail” showing the mail setup, and “web” showing the subdomains. Or maybe they really should be listed seperately - I can’t decide. (For reference here is my live listing of domains - where the problem becomes obvious. Note that you’ll not have permissions to see my server stats…) So I need to decide how to handle this - I also need to be "inspired" by OSWD. I think the plan right now is to store all data in a database, then regenerate the plain text configuration files on demand. That should give me the best approach - if I restrict myself to running HTTPS on a custom port and avoid the ability to modify the files by hand… Feel free to tell me this is misguided, or inspired. Although I remember the last time I talked about control panels and I’m still not seeing a good solution. I want to be able to:

David Nusinow: It's Not A Crisis Until The Java Guys Actually Take Over

Lately I keep seeing people talking about a "crisis" in the ruby community due to the fact that rubinius, jruby, and YARV are all being written. There seems to be some subtle hand-wringing under the covers about divergence between these implementations. The ruby language is in pretty solid form, with a few little problems that should be cleaned up for the 2.0 release, so I don't think it's very likely that any divergence will be particularly damaging.

More importantly though, I think it's a sign that ruby is alive and well, given that there's three different projects that are working to provide an improved implementation for a language that was largely ignored until two or three years ago. Plus the benchmarks for YARV (which was just merged to the ruby trunk) look pretty hot.

Unfortunately for the doomsayers, most of what I've heard and read from the actual ruby community about all three of these projects has been a giant ball of excitement. I think it's really sweet that we'll have ruby running on the JVM, on a sexy Smalltalk style VM, and also a significantly sped up classic interpreter. Now all I want for Christmas is more time to actually write the rails site that's been pounding away inside my head like Athena.

Christian Perrier: l10n NMU campaign, second edition

I started today the winter campaign of l10n NMUs, after a long and cool autumn campaign. This time, this is more a kind of BlitzKrieg: No QA improvements this time, only pending debconf l10n stuff. Targets are chosen among pending French translations. After all, it's fair that the effort benefits the team I'm part of for years, right? So, fellow developers, if you have pending French l10n bugs, better watch your back...:-)

Enrico Zini: latex-beamer

Some latex-beamer tips Using with source-highlight source-highlight's latexcolor output uses colors that are not available in latex-beamer. Normally one would have to include them with \usepackage[usenames,dvipsnames] color , but under latex-beamer it does not work. One solution is to include the relevant colours from /usr/share/texmf-tetex/tex/latex/graphics/dvipsnam.def:
%% For source-highlight's output
% This unfortunately does not work
%\usepackage[usenames,dvipsnames] color 
% Therefore, I will copy them from /usr/share/texmf-tetex/tex/latex/graphics/dvipsnam.def
\DefineNamedColor named Blue            cmyk 1,1,0,0 
\DefineNamedColor named BrickRed        cmyk 0,0.89,0.94,0.28 
\DefineNamedColor named Brown           cmyk 0,0.81,1,0.60 
\DefineNamedColor named ForestGreen     cmyk 0.91,0,0.88,0.12 
\DefineNamedColor named Purple          cmyk 0.45,0.86,0,0 
\DefineNamedColor named Red             cmyk 0,1,1,0 
\DefineNamedColor named Black           cmyk 0,0,0,1 
\DefineNamedColor named RoyalBlue       cmyk 1,0.50,0,0 
After this is done, including a source is just a matter of:
# In the shell
$ source-highlight -f latexcolor filename.ext
# In the latex-beamer frame
\input filename.ext.tex 
Proportional columns In the latex-beamer documentation I have only seen columns with fixed sizes. \textwidth luckily works:
\begin frame 
  \frametitle Correzione dei compiti 
  \framesubtitle Il programmino da Fortran. 
  \begin columns 
  \column .5\textwidth 
   \tiny
  \input fortran/01-inizio.f77.tex 
   
  \column .5\textwidth 
   \tiny
  \input fortran/02-inizio.py.tex 
   
  \end columns 
\end frame 

Enrico Zini: suspend-on-asus-u5f

Suspend on ASUS U5F I have suspend. The acpi4asus developers now added support for the U5F. I pulled their code and now I can control the screen brightness writing numbers into /etc/acpi/asus/brn. I also discovered that the problem I had on suspend was that the laptop wakes up with some weird brightness setting that causes the light not to turn on. Writing to /etc/acpi/asus/brn I can now manage to turn it on again, and I added trivial script that save /etc/acpi/asus/brn on suspend and restore it on resume. \o/

Ian Murdock: Open source: Predatory?

Gordon Haff: “Imagine, if you will, that it s the late Nineties. A certain software company based in Redmond, Washington has recently released Visual Studio 97 thereby bundling together many of its development tools for the first time. Now imagine that the company decided to release those tools for free. What do you think the general reaction would have been? Applause for Microsoft s generosity? Or widespread condemnation for using its market power to make such a transparently anti-competitive attack on other makers of development tools?” Thought provoking observations. Note, however, that “open source as competitive weapon” isn’t limited to large vendors—it works equally well (and, in many ways, better) for the upstarts (see Red Hat, MySQL, JBoss, etc.). In fact, it works so well for the upstarts that even the upstarts have upstarts (see, e.g., Canonical). Can something that levels the playing field so dramatically be called predatory? Predatory has one-sided, unfair advantage connotations. In open source, it goes both ways. With Eclipse, it was IBM doing the disrupting; but with MySQL and JBoss, IBM is on the other end of it. At the end of the day, the real winner is all of us—after all, who can argue that the state of IDEs and middleware isn’t better today than it’s ever been?

Russell Coker: LCA talk

This afternoon I gave a talk at the Debian mini-conf of LCA on security improvements that are needed in Debian, the notes are online here.

The talk didn't go quite as well as I had desired, I ended up covering most of the material in about half the allotted time and I could tell that the talk was too technical for many audience members (perhaps 1/4 of the audience lost interest). But the people who were interested asked good questions (and used the remainder of the time). Some of the people who are involved in serious Debian coding were interested (and I'll file a bug report based on information from one of them after making this post).

I believe that I was quite successful in my main aim of giving Debian developers ideas for improving the security of Debian. My second aim of educating the users about options that are available now (with some inconvenience) and will be available shortly in a more convenient form was partially successful.

The main content of my talk was based on the lightning talk I gave for OSDC, but was more detailed.

After my talk I spoke to Crispin Cowan from Novell about some of these issues. He agrees with me about the need for more capabilities which I take as a very positive sign.

David Moreno Garza: Ulteo

I’ve been hired by the Ulteo project to work on packaging and releasing matters. I’m pretty excited ;-)

Brett Parker: Random Interesting Things of the Day

iPhone news So, looking around at what's happening in the media we seem to be getting a fair few non-overly impressive insights in to the iPhone (not least of all the fact that Apple haven't secured the name, and that Linksys have had the iPhone name for a while now - see http://www.linksys.com/iphone/ for the real iPhone, the one that does exist, the one that's already in the market place...) So, back to Apple, people (being people, and therefore incredibably fallable ;) have created iPhone themes for current mobiles, and they're being hunted down by Apple's lawyers, that can't be good publicity, surely? I thought imitation was the sincerest form of flattery... tut, bad evil corporate giant! According to Matthew Lynn on Bloomberg's opinion pages, """To its many fans, Apple is more of a religious cult than a company. An iToaster that downloads music while toasting bread would probably get the same kind of worldwide attention.""", which seems about true to me - the rest of the article is worth reading, gives more insite in to why he thinks it will fail, and is quite an interesting article. Health There's a story in the Guardian about Low cholesterol levels linked with higher risk of Parkinson's disease - so we'll continue the Monday trip to the Cafe then, and have us a nice lunch of bacon and eggs, in the name of health, obviously. From NewScientist, we get the news that being bilingual delays the onset of dementia, so maybe it's time I got round to learning a language that isn't English... I wonder if knowing a reasonable number of programming languages and their syntax works... Someone should do a study in to that. Random Tech And just for more fun, lets just have a bunch of interesting links Miniature jet engines could power cellphones, Silicon 'Lego bricks' used to build 3D chips, Gravity gets a quantum boost. And that's that... So, we'll end that there, tonight is, as with every Monday, the pub quiz over in the Hop Poles in Brighton. Should be very silly, and fun.

Miriam Ruiz: Playing with usplash

I’ve been experimenting with usplash for some weeks, and it seems quite powerful. I decided to try with the latest version of the program (0.4), from Ubuntu’s repositories, which doesn’t seem to be backwards-compatible with the one that is currently in Debian (that is, 0.3e). The main idea of a theme in usplash seems to be embedding all the needed information into a shared object, including the images and the fonts, and coding the themes in plain C. This, which might seem terribly scary for non-coders, gives a lot of freedom for creating new effects in a certain theme, which is something I want to experiment. The low-level graphical fuctions use bogl (Ben’s Own Graphics Library) and svgalib for showing, and the rest of the code is built above them. Even though I like the overall design of the project, it doesn’t really suits my needs for what I want to try. I need to implement more low-level functions for certain effects, and I’m also considering doing some changes to optimize memory used. Right now I’m doing the graphical effects in a pixmap and, right at the end, dumping that pixmap onto the real screen, which makes me adapt what I want to the canonical code of usplash, but I guess at some point I might need to make deeper changes in the code, and I might need to change the insides of usplash.
I don’t know if something useful can come out of this, or if the results will be only aesthetic, but I’ll put my code available somewhere as soon as it is ready for inspection, so I can get some comments on it. As always, ideas are welcome.

Erich Schubert: Smoking kills

Nothing new, but fun: Sometimes, smoking can be a very bad idea. Darwin award 2006.

Julien Danjou: Kicking out Web spammers with DNSBL

Every project has its story. Every war has its winner, and its casualties. They were 20 millions men, fighting for their freedom. And you'll never know their story. Because during last week, I was looking why my Web server was so heavily loaded. And I discovered that my blog was attacked by spammers trying to post comments. They were stopped by a great plug-in named spamplemousse, which use spam keywords and DNSBL to drop spam comments.
However, this plug-in is written in PHP, like the rest of my blog, so it loads Apache and MySQL in a way that is no more acceptable: the page have still to be rendered for this !@#$ spammers. Consequently, I decided to write a Apache 2.x module which will just drop a 403 Forbidden error page in the spammers' head using DNSBL servers. Here it is, and it is called mod_defensible. I'm using it since 3 days now, and I got some pretty interesting result and less load on my Web server, so c'est tout bon[1].
Notes [1] It's all good

Martin F. Krafft: Status of my book

The wait for Debian sarge was long and painful for me, because I was trying hard to synchronise the publication of my Debian book with it, and much to my own surprise, the book was done before sarge was ready, and I had to wait. In the end, the tome left the presses only a little more than 24 hours after our release managers announced Debian 3.1 in July 2005, bundled with the second release of the official DVD, 3.1r0a. Since then, the volume has been reviewed numerous times and praised by many readers. I was especially honoured to find the title in the top ten list of most favourably reviewed books by one of my most respected reviewers, Richard Bejtlich. We are now approaching the release of Debian etch, version number 4.0, and a lot of stuff has been added or changed in Debian. I have received more than 150 emails since the etch freeze announcement, asking about an updated edition for etch, and while I am most gratious for all the feedback (and can't deny a certain amount of pride), I have to disappoint my readers, at least for now. I have not had the time to update my text, and thus cannot synchronise a new edition with the release of etch. There are a number of reasons: first, shortly after the publication of the original edition, I suffered from several physical problems, which in the end all pointed to severe muscle deformations in my back from years of wrong posture sitting at the desk. Now, one and a half years after the first symptoms appeared, I've been through numerous therapies and even though the net result is positive, I am not over the problems yet. Thus, I still try to minimise the time spent on a computer, and getting a second edition of the book out would achieve quite the opposite. Second, I am busy with my Ph.D. research, so I cannot work on the book much. It has been suggested to me to allocate minimal resources to an update, getting the text ready for etch and postponing new content to another edition. However, I would feel uncomfortable doing that. On the one hand, I'd much rather see my work be relicenced under a Creative Commons licence. So long as that is not possible, I'd rather not make people spend money without getting much in return. My book's to-do list is growing and there are many new topics which I would like to cover in the next edition (for instance: fonts, xen, more packaging topics, including modules, kernel patches, pbuilder, and packaging with version control systems). Thus, with the release of a new edition, I want to provide not only an updated version, but also one with enough content to give a fair return for people's money. I have always thought that updated-but-no-new-content editions are ripping the customers off, so I don't want to go into that direction. I am thus currently looking at the relase of the second edition together with Debian lenny, the Debian release following etch. I am looking forward to much more feedback between now and then. NP: Solar Project / Force Majeure

Lucas Nussbaum: ruby & native threading

Stefano Zacchiroli said:
To balance this, according to my first read of Ruby’s threading capability, it’s my impression that not only at most one thread can execute Ruby code at a time (limitation shared by OCaml, due to the non-distributed nature of mark and sweep garbage collectors), but also a thread blocked on a syscall will block all other threads to run. Dumb Ruby threads (but I still hope I’m wrong …)
This is not totally true:

Evan Prodromou: 24 Niv se CCXV

Last minute note: I mentioned some discussions about advertisement on Wikipedia; I sadly missed a conversation in which Jason Calcanis pointed out that Wikipedia leaves $100M on the table, and the response from Jimmy Wales who has his own take on advertising and Wikipedia. Interesting discussion. tags:

Joey Hess: debootstrap's 6th anniversary

Sometime this month, debootstrap turns six years old. I was thinking about this last night as I tweaked a new xen instance that the hosting provider had installed Debian on using debootstrap, while in the other window using a debootstraped system to investigate a weird apt premissions problem. All very standard, and between this kind of thing, installing Debian with d-i, and building packages with pbuilder, a lot of us would be lost without debootstrap now. The story of exactly how it came about is interesting, though it's blurred by time in my memory. Maybe AJ will correct me, but I think it went something like this: Debootstrap is one of my favorite examples of a tool that ends up being much more general purpose than was first envisioned (at least by me; AJ may have known all along). The collection of stuff that has grown up around it and the way it's changed how both Debian developers and users work is pretty impressive.

14 January 2007

Uwe Hermann: Upgraded my website to Drupal 5

I have upgraded my website/blog/podcast/photoblog/linkblog/whatever to Drupal 5.0-rc1 today (I'll skip 5.0-rc2 for now and wait for the final release of 5.0 for the next upgrade). The upgrade went quite nice, even though I had to upgrade several modules and port quite a bunch of custom hacks I had on my old (Drupal 4.6) site. I first upgraded to 4.7, then to 5.0 (as is the recommended procedure) on a test-site, and after figuring out how to fix or work-around all the issues that appeared, I upgraded the live site. I password-protected the site during the upgrade, that's why it wasn't available for a while today (and caused some problems on Planet Debian it seems, sorry for that!). New features you might enjoy: If you notice any bugs or problems with the site, please let me know.

David Welton: Stuff To Do updates

After two weeks of being operational, Stuff To Do has been a big success at work, because it lets people keep track of what they're doing with a minimum of hassle. Truth be told, I've never been that wild about "productivity" software, but I'm pretty happy with the system I've created. I really have attempted to make something that stays out of your way as much as possible. That has involved making some choices - for instance, instead of buttons to start or stop time tracking for a given task, Stuff To Do [aside: it would be much faster to write 'STD', but I think that might not be such a good idea in terms of marketing] uses some javascript to see if you've moused over (or blurred, or typed, or done anything, basically) to the window where it lives in the last 15 minutes. This will certainly reduce the accuracy a little bit, and if you get 'in the zone' and don't hit the browser for a while, it won't count your time, but all things considered, this is better than having to hit a stupid button every time you do anything, because you will invariably forget to hit the button, and leave it going overnight, and then have to go back and manually edit everything, and.... yuck. I also have some ideas about how to further improve the javascript in order to make it even easier to track whether you're active or not (with your consent, of course!). In any case, I've fixed up a number of small bugs, and added a few new features: Some other things I'm considering for the near future: In the process, I also happened upon Active Merchant, which is some cool code by the fine folks at jadedpixel.com, which was quite handy for integrating Paypal. Being somewhat of a perfectionist, I found a few things that I wanted to do just slightly differently, and sent in a patch.

Next.

Previous.