« July 2005 | Main | September 2005 »
August 30, 2005
Understanding Röyksopp
The new Röyksopp album came out earlier this year. A good one, again.
Admittedly, the stakes were high after Melody AM three years ago. Back then, they were thought to be innovative and chic. It's music that reminds me very much of the coolness of our artsy techie MLE back then.
Röyksopp's new album The Understanding is solid. It's not terribly innovative compared to what they did with their first one. Instrumental songs like Alpha Male are very much like what Jean-Michael Jarre did twenty years ago. But what was great back then doesn't have to be bad today (just boring).
Their title track Triumphant is on hot rotation at KCRW right now, and it sure deserves that spot. Someone like me grooves as appropriate.
Get The Understanding. Röyksopp is cool, even though we all have learned to spell their name without googling.
Posted by dr at 10:40 PM | Comments (0) | TrackBack
August 26, 2005
Capital Car and Van Hire, Edinburgh. My rental-car nightmare.
Lisa D. works at an Edinburgh rental car company. Her job: handing out funny contracts to customers and getting them to sign statements like "Lessee and/or third parties agree to allow Lessor to compute and debit additional charges From Credit or Debit Card, if that is the form of payment used. Additional Charges include damages, extensions, penalty charges, administrations charges or any other charge sought to be imposed."
My Legalese may be not up to scratch, but what I read out of this is that you're allowing them to charge pretty much anything they want to your credit card the instant you put your signature under that contract.
Well, it's always like that. They make you sign funny things, and nobody would ever suspect that the company is going to try to use that against you. At least you're not very inclined to think so when you have 10 people lined up to help for the move, waiting for you to come back with the moving van, which is about the last one available in town.
I and a number of other customers made the mistake of handing their credit card to Capital Car and Van Hire, Haymarket, Edinburgh. What happened in my case is only half as funny as it sounds. The story begins on a bright Saturday morning, on which I was to help my office Mate Myrosia to do her move, just before going on a vacation. The truck I got from Capital Car was just fine and did the job. But upon returning the vehicle, Lisa came out to inspect it as is usual. To our surprise, she claimed that the "bumper was out of alignment". WTF? I couldn't see anything wrong with the bumper, neither could Myrosia. Nor did I know that bumpers have to be alignment, for god's sake! Of course there was nothing wrong with the car - it hadn't been involved in any accident.
Turns out, Capital Car just needed a pretense to keep our deposit of £200. But what's worse, without a warning Capital Car charged £700 more to my credit card the same afternoon. The card was maxed out by then, and I noticed a few days later.
What to do?
I guess you know those lawyer jokes that people make at parties, and which David makes at parties full of lawyers (without getting kicked out - yet). I tell you what: none of them is true. Every single lawyer I've met was a exceptionally nice. They are well-educated, forthcoming, friendly, and what's most important, whenever I had to see a lawyer for some legal issues, he told me that my view of the legal situation was spot-on.
That's what happened in my case, too, when we saw an Edinburgh-based lawyer. Funnily enough, the good man could guess what rental company it was. He regularly gets clients that complain about Capital Car, he said - usually they have similar stories to tell. As we find out later, a co-worker of ours lost a £200 deposit to Capital Car under very dubious circumstances. And a Dutch couple, on vacation up here in Scotland, are still a couple hundred short even though they won a court case against Capital.
Months have passed, since, and at this point we have some of our money back. Since I didn't sign off on those £700, I could initiate a charge-back through my credit card company - that is, effectively reverse the charges. This cost me hours of phone calls and work, since they want a detailed written statement.
So what's next? I call it fraud what's happening there, but whether that is legally proven to be the case probably depends on some people at the attorney general's office to invest time. Also, there's a trading standards committee that ought to be investigating the case of Capital Car and Van Hire. And of course, we can (and possibly will) go to court. We won't get the money back - we would just get it from a corrupt company and take it to lawyers and the government for court fees. But that'll be alright. At least we'll get a warm fuzzy feeling from having done the right thing.
Posted by dr at 10:04 PM | Comments (22) | TrackBack
August 24, 2005
Helicopter in a tunnel, my ass.
Several airplane accidents in the past weeks, my recent attempts to land Gulfstream Jets in variable crosswind at EGPH Edinburgh airport (well, in the simulator) and the German re-elections have taken their toll.
Last night I dreamt of getting on a small helicopter, accompanied by (or accompanying) the German chancellor / prime minister Gerhard Schröder. The pilot successfully steared the chopper through narrow tunnels and under bridges.
In hindsight, we had better taken a car. Hitting something with the rotor is very much unadvisable and causes devastating damage.
Posted by dr at 10:26 AM | Comments (0) | TrackBack
August 22, 2005
Aquamacs is not just an ego-trip.
OK, this is going to be a technical and maybe political read for people who haven't had a good taste of the free software development world. It's about what's been going on with regards to Emacs (on the Mac).
Andrew Choi, the man who brought GNU Emacs to the Mac, believes that "free software developers are in it for Ego, not Freedom". He's right and wrong, both for the wrong reasons.
Andrew takes great exception at the fact that recently, Emacs enthusiasts have begun offering their own Emacs distributions.
They tweak and add code in order to save people time configuring and running Emacs on their system. Sometimes the changes are small, sometimes they're not. Aquamacs, my and my collaborator's baby, is one of these distributions. In Andrew's view,
Distributions are given fancy names like Enhanced Carbon Emacs, Aquamacs, YACED, etc., but they are really just packaged Carbon Emacs. I suppose packagers think they're entitled to name a software after making a few small changes. We aren't talking about big Linux distributions here people.
Aquamacs Emacs contains over 5100 lines of elisp code and about 1200 lines of patches to the C core of Emacs, 99 percent written by me.
The Japanese Carbon Emacs Package due to Seiji Zenitani and others contains a lot of carefully written code, too, including a fancy Net-Install feature.
In addition, Aquamacs comes with a number of third-party packages pre-configured. Finding, installing, configuring these packages to provide a useful UI consistent with Mac/Windows environments is of substantial added value to our users.
I'm well aware that the changes are small compared to 30 years of Emacs development, and Andrew Choi's gigantic effort porting GNU Emacs to MacOS and later to OS X. But in terms of what the user gets to see, Aquamacs already represents a departure from certain Emacs usage paradigms. It needs a different name.
Think of the one-buffer-one-frame philosophy (GNU Emacs traditionally displays files in the same big window ("frame"), rendering the OS's beautiful windowing system useless. It has it's own windows-inside-the-frame, pre-dating real GUIs.) Think of OS-standard-confirming shortcuts. Finally, it's a different philosophy: we see Emacs as an application and not as an entire work environment, which seems to be the predominant view among old-style Emacs enthusiasts who even like to browse the WWW in Emacs.
The use of imprecise words like "enhanced", "based on CVS Emacs", "Aqua", etc. in descriptions of distributions implies more improvements than have actually been made.
Maybe we're from different planets. But the Enhanced Carbon Emacs is enhanced. Just to give an example: before the advent of those distributions, GNU Emacs developers - Andrew including - didn't find it necessary to provide off-the-shelf fonts with anti-aliasing enabled. Instead, a font menu was (and still is) present in the Emacs provided by the Free Software Foundation that offers some 12 fonts, none of which can be used to display one's texts. It doesn't work, and when it does, it's not ergonomic.
Of course, one has different priorities. But Andrew's opinion clearly shows that many Free Software developer are indeed egotistical. They don't care much for the user. They own habits and preferences overrule whatever standards are established for the UI of applications.
That's why we created Aquamacs Emacs.
People package and redistribute Carbon Emacs typically sell and promote other products at their websites.
First of all, there's nothing wrong with selling and promoting products. People have to live from something. Secondly, neither Aquamacs, nor ECE, nor Yaced sell anything else. Not even Aquamacs' original home at WordTech sells things on the same website. Instead, co-inventor Kevin Walzer sells poetry books as a publisher, and most likely not to Emacs geeks.
Of course, I'd like to at least make some money to pay for the website. But it seems like that's not going to happen. I don't give a flying f... about a potential employer in the Linux world, who might one day pay me to hack elisp. I do research and a maybe a bit of journalism for a living, not Emacs hacking. Consulting jobs would be great, but I haven't had an offer even after thousands of Aquamacs downloads. To be quite clear: it's charity work that I'm doing. And if I can gain a bit of respect, then that's a very meager return-on-investment.
I got involved for two reasons. I wanted a decent input method including syntax coloring for Prolog programs, and BBEdit didn't offer that. I didn't want to use plain Emacs because it represents annoyances to someone who is very much used to a certain user environment. Secondly, I wanted to learn Lisp. Thirdly, I wanted to know what other people would think of a mac-ified Emacs. The first two objectives were reached. The third one was a question that thousands of people have answered by downloading Aquamacs. A substantial number of people use it daily (I track these things!). Enough motivation.
The disappointing fact lies in something else that Andrew once wrote. There is no Bazaar in Free Software Development.
The code was written by myself, and I took some prefabricated packages. The manual was written by my collaborator. And Emacs was Emacs. That's it. The others are users with great technical interest -- none of them has actually contributed more than a few lines of sample code. I am unable to keep implementing features - I have a day job.
There are people who mislead and those who plain lie! My favorite example is a website that lists Carbon Emacs as one of the programs its owner have "ported or written". Huh? If he did that, what did I do?
Well, that's where Andrew is correct. People lie. Welcome to the real world. I do get annoyed by folks that offer a plain compiled Emacs without extras as a binary and claim that they provide something special.
New mailing lists were eagerly established to discuss the Carbon port instead of people being sent to join the main discussion at emacs-devel. Of course, a few more people can now become "experts" on these new lists.
Well, truth of the matter is, we discuss Aquamacs user-related matters on a long-established Emacs-on-OS X mailing list - not on a new one. With regards to bug reports, users started reporting issues to the Emacs-Bugs mailing list as soon as I had fixed the bugreporting function (it began to actually work on a Mac after years of swallowing the bug reports without notice!). But what happened is that Emacs inventor Richard Stallman got (understandably) annoyed with bugreports that related to Aquamacs-only issues. So he asked me to direct users to our own mailing list, which we did.
We are and we will be in competition with the main Emacs project. It's a competition for more users, and for better user interfaces (because that's what an editor is about). But diversity and competition are good things, especially in a system where people can feed on other people's code.
A significant number of people who make no or negligible contributions to Emacs/XEmacs code hang out on mailing lists and newsgroups and act like real experts and bully novices and other posters. People who know even less then chime in to make matters worse!
User support is something important, and it adds value to the community. And it's even more important that the experts invest their time into developing the project rather than answering beginner's questions.
I'm no Emacs expert (in my view), but to Aquamacs users, I am. It's all relative.
Aquamacs has already had a small impact on GNU Emacs development. We contribute back to the main project - in Choi's view, Emacs distribution maintainers seem much less concerned with adding their changes to the main CVS.
I contribute code where I think it's useful and where I think the maintainers will accept it. I have reported countless bugs (takes time, too!). I will continue contributing code to the main project. But I am not willing to enage in week-long debates about silly little things that seem just plain obvious to someone coming from graphical user interface based systems. If people want the code, they can have it. It's there, and I'm happy to help. But I'm not happy to take a stand in long pseudo-political discussions, or bash people on mailing lists whom I respect way too much for their long-standing work. Andrew Choi and Richard Stallman are two of them.
Posted by dr at 11:29 PM | Comments (1) | TrackBack
August 20, 2005
Blog searches
One of the most common use of search engines is to look for oneself: "About one in four Internet users has typed their own names into a search engine", finds a recent study conducted by an American research center. Men are more likely to check up on what others think about them.
Not so on my blog: strikingly common search key words are the names of women that played and play a role in my life. What do I write about them? Disappointingly, it's usually not much that is to find here. No nasty good-bye-letters, no naked people (check back later!).
Other searches seem to look for projects that I once published on the Internet. A lot of people look for Heiner Müller, an important German dramatist about whom I've written a paper Heiner Müller im Spiegel der Nachrufe in 1996. Another paper people look for was just a little class presentation about The Death of Roland Barthes' author, titled A Short Absence of the Author. Always good to know that I once used very different methods of looking at knowledge and language than through the directed graphs and trees and constraints and Markov models that I'm more familiar with nowadays!
Sometimes, people get to my page because they think I've written about them, and then they search this blog for their names. The reason for this is that I have a few lists of bogus names on the web. That was a fun project to see whether how easy it is to generate such names (using publicly available census data). People think they've found themselves, because their name popped up randomly. By the way - each of these names I've linked to a randomly generated, but seemingly realistic e-mail address in order to "Spam the Spambots": these entries generate a lot of crappy data in the big databases of people that send out unwanted advertisements per e-mail. Such databases are generated automatically from random web sites - such as this one.
Posted by dr at 8:16 PM | Comments (0) | TrackBack
August 15, 2005
A growing fan base for Aquamacs
Great news: Aquamacs, my Emacs-for-Mac OS X-project is gaining new fans every day. I ran some statistics (I like stats!!) on the data I've been collecting from the application's "version check" feature, where it connects to our server every three days or so to see if a new version is available. We take note of that and log the installed version and a few other things - strictly non-personal info, of course. Most applications do this - Aquamacs is open source, though, so we're being open about it.
The results are very motivating. About one third to one fourth of all people who try it out stick with it. The latest release (0.9.4) generated quite a bit of interest. We have users from 64 different countries, and as the diagram below shows, we're getting more every day. Even though I can't devote much time to it because I'm working on interesting research, the next version will be even more reliable, snappier and have a few new functions and a better menu structure.
By the way: Aquamacs is a joint project with Kevin Walzer (Wordtech, Ohio), who has written awesome documentation for it. We're looking for other co-developers, by the way!
Posted by dr at 10:39 PM | Comments (0) | TrackBack
August 12, 2005
Ant roboting: Lots of little guys trying to become a smart mob
I haven't written much lately here. The Silence of the Lambchops. "Boy, you better have damn good reasons for that!" Yes, Sir, I have reasons.
a) slightly different priorities (writing a research proposal) and b) attending conferences. IJCAI for example.
IJCAI is one of the largest conferences in Artificial Intelligence research, and it was fun meeting people in the field and finding out what other researchers are up to. People in "Swarm Robotics" build miniature robots that can team up to achieve simple tasks. If a few of them break down: no problem! Among the questions of interest are: how do they communicate in the network, when they can only transmit to their neighbors (lacking a strong battery)? Just like ants or bees or the Internet, they can pass on messages in a redundant information network. How do the agents assume roles and commit to their tasks? How does a distributed system integrate redundant information and reason under conditions of uncertainty?
These are quite practical problems: it's not that trivial to actually instruct them to do a specific task as a team, such as delivering drugs right into a tumor somewhere in the human body. An excellent tutorial about collaborative multiagent systems, given by Barbara Grosz (from Harvard, she's highly respected for her work on human dialogue and discourse coherence), Charles Ortiz (SRI) and Milind Tambe (USC CompSci) helped me understand better what the principal issues are. Too bad that the practical Swarm Robotics workshop could only show somewhat simplistic applications and systems. I wonder if the robotics researchers would be better off simulating little robots in software form...
Posted by dr at 10:17 AM | Comments (0) | TrackBack