Films seen in March 2010

I’ve seen the following this month:

The list features a lot of foreign language films. Waltz with Bashir, Khadak and Gomorrah are respectively Hebrew, Mongolian and Italian language films, and The Wind That Shakes the Barley is  an Irish film. It’s not that the American films lack quality in any way, but it’s liberating to see something different and evade the American cultural domination.

Rachel Getting Married is a very good film. If I had to summarize the film in a proverb, I’d say ’once a thief, always a thief’. In it’s essence the film explores the question if we should keep condemning persons for past sins or forgive them. It does that skillfully, and ends happily with a wonderful wedding day party which is a wonderful experience. Vicky Cristina Barcelona is a good comedy, it’s an excellent advertisement for visiting Spain. I’m still not convinced that Barcelona in particular would interesting to visit though. Into the Wild had a surprising ending for me, if you want to be surprised too, don’t look it up on Wikipedia or read the book. I keep thinking how I would have dealt with the river with the strong current. Assuming there weren’t any safer places to cross it, I think I would have risked crossing it by swimming, but I’m saying that with the wisdom of hindsight.

The Last King of Scotland was the only film I didn’t rent, it was broadcast on TV (not sure if it was in March). Whitaker does a awesome job depicting a madman like Idi Amin, and definitely deserved the Academy Award he received as best actor. The film got me thinking about the authoritarian regimes and low quality of government in Africa in general, is it all caused by European decolonization, or are the Africans themselves too blame as well? It’s sad to see that so many people followed Idi Amin and executed his orders for their own gain, even though they knew Amin was mentally ill and all he did was bring misery to Uganda. Of course ‘civilized’ Europeans also suffered from this in the first half of the 20th century, and considering that Europe had a longer history of development and never experienced decolonization the Europeans are the ones who deserve to be blamed most of all.

To keep true to the theme of proverbs, when I was watching The Wind That Shakes the Barley I had to think of an aphorism written by Nietzsche. ‘He who fights with monsters should look to it that he himself does not become a monster. And when you gaze long into an abyss the abyss also gazes into you.’ The British are the oppressive monsters, and in fighting them the Irish lose their ideals which they were originally fighting for.

Waltz with Bashir is an eye catcher because it is an animated film. Do not let the fact that it is animated fool you, it’s far from child’s play or rubbish. The animation is excellent. The film is a journey towards the past, the truth which is reached at the end of the film. The distinction between dreams and reality is fluent in this film without being confusing, unlike the next film.

Khadak was a challenging film to see. The distinction between reality and dreams was blurred, which makes it difficult to make sense of the film. It demands much interpretation, it makes a heavy use of symbolism and long silent scenes. The cinematography is awesome with shots of a huge, flat and empty Mongolian steppes covered with snow, depicting only a shepherd and his horse. Certain shots are very impressive, such as the old nomad who doesn’t want to break with his traditional way of life sitting on a chair, with two government officials lifting his chair to drag him to a truck so he can be transported to the ‘civilized’ world. To summarize the film I’d say that it is criticism on the encroachment of the modern world on the traditional nomadic way of life in Mongolia. The first half progresses very slowly, I got frustrated and had to convince myself that I wanted to see it to the end. When I finished the movie, I was glad I had done so, I felt rewarded.

Gomarrah was a disappointment. It’s not a bad film, but it’s not executed good enough. The film uses an ensemble cast, which is not a problem in itself, but it does become a problem if the cast is large and the story lines become confusing because there are too many and they ‘switch’ too often. It was difficult to understand what was going on, my father was watching too and read the book the film is based on. He easily understood that the film depicted illegal toxic waste dumping and illegal garment factories, I didn’t understand that so easily. After having seen half of the movie my father got bored and wanted to see the sports news, I had to agree with him that it the film wasn’t really going anywhere. The next day I watched it to the end alone. In the second half the murdering begins and many people including part of the cast end up dead. The film doesn’t feature much of a plot, it mostly depicts mafia members pursuing all kinds of illegal activities, the bottom line is that the mafia are bad guys. This film – even though it is quality – didn’t deserve to be on some critics top ten lists for 2008, which was the reason it was on my list of films I wanted to see.

Problems with power saving on Kubuntu

In my previous post on KDE and Kubuntu’s power consumption I mentioned that Kubuntu and/or KDE’s power consumption seems rather high. Since then I have investigated the problem. All the testing I’m describing here is still done with my Acer TravelMate TimeLine 8371 with the latest 1.18 BIOS, but with the latest Kubuntu Lucid a.k.a. 10.04 and the 2.6.33 kernel from the Ubuntu kernel PPA.

I’ve posted on the KDE forums to ask why I experienced such a high power consumption. With the help from others, I figured out that Kubuntu doesn’t do a good job at power saving because of some bugs. Running PowerTOP on the 8371 gives me the following suggestions to reduce consumption:

  • Enable USB autosuspend for non-input devices by pressing the U key
  • Enable SATA ALPM link power management via: echo min_power > /sys/class/scsi_host/host0/link_power_management_policy or press the S key.
  • An audio device is active 100.0% of the time: hwC0D2 Intel G45 DEVCTG
  • An audio device is active 58.6% of the time: hwC0D0 Realtek ALC269
  • Disable the unused WIFI radio by setting the interface down: ifconfig wlan0 down

The suggestion for USB autosuspend is affected by bug #136549. Apparently the USB autosuspend is already enabled by the default and PowerTOP shouldn’t suggest changing it at all. SATA ALPM link power management makes a serious difference in power consumption, but for some reason only PowerTOP can change it. The power saving script I created or a manual command doesn’t work, this bug is described in bug #427925. For PowerTOP’s suggestions regarding enabling HD audio power saving because the audio devices are 100% of the time or less active (while no sound is playing at all) I filed bug #521424. Trying to enable HD audio power saving with my power saving script doesn’t work, and PowerTOP can’t do it either because it keeps repeating the suggestion to enable it many times even after you have done so. I believe the phenomenon of KDE asking to remove an audio device, as described in my post on the KDE forums, and here, here, here and here. I’ve seen this question occur a lot right after logging in to KDE, but recently since I’m using Lucid and 2.6.33 I have never or very rarely seen it, I’m not exactly sure. I think bug #509708 is related to this. The last suggestion to disable the unused WiFi radio doesn’t work either, it comes up when I disable the WiFi through the hardware button, but PowerTOP also keeps repeating this suggestion after you hit the key to apply the suggestion.

Problem is that the Kubuntu team isn’t making any effort to fix this bug regarding the audio devices being active 100% of the time or less. It’s a Kubuntu specific bug it seems, but so far my bug report hasn’t even been triaged after a month, even after asking attention for it on the kubuntu-devel mailing list. Being a volunteer myself, I understand that many who volunteer don’t have much time on their hands, but some of Kubuntu’s developers are also on Canonical’s payroll. Sometimes even relatively trivial bugs which are easily fixed take a year or longer to get solved, for example bug #291048 and bug #338285. If it takes months before anyone even responds to a bug report or if there is no response at all, it is demotivating for bug reporters to keep reporting bugs.

If these bugs concerning SATA ALPM link power management and HD audio power saving are still present in the stable release of Kubuntu 10.04 LTS (Long Term Support) I’ll start looking for another distribution which ships KDE 4 and isn’t affected by the problems, Chakra draws my interest. The long battery life of over eight hours was one of the primary reasons for buying my TravelMate 8371 notebook, so I’m not settling for five hours because my favorite distribution, Kubuntu, doesn’t do power saving properly. I’ll make sure to compare the power consumption between Ubuntu, Kubuntu and another KDE distribution again when Lucid Lynx is released at the end of April.

Combating Microsoft’s dominant market position or monopoly

Months ago I wrote about Microsoft’s business practices and the difficulty to obtain a student license of Windows 7 for a low price. It turned out that isn’t possible to upgrade to Windows 7 from Windows 2000, so I can’t buy Windows 2000 for € 15 and then buy the upgrade version of Windows 7 Professional for students for € 53,00. I also discovered that the Windows 2000 license offered for € 15 on the most popular Dutch advertising website Marktplaats was an illegal license, because the seller admitted it. If you search on Marktplaats for offers on Windows 7 Ultimate you’ll find many offer illegal versions for under € 20, while it’s at least € 155 for a legitimate OEM-license. I wonder why Microsoft or Marktplaats itself doesn’t crack down on these sellers there, because I’ve been seeing such offers for months.

I can’t even buy an OEM-license of Windows 7 Home Premium for € 84 because you can only buy that with a new computer (Microsoft Netherlands condones it in practice, but in their e-mail communication with me they insisted that it isn’t allowed). I’d have to spend € 157 on a retail license for Windows 7 Home Premium. Sorry, no deal for me. I’d like to buy a legitimate product, but € 157 is very unreasonable. I’d be willing to pay € 50 at most.

If we apply the average (because OEM-licenses sold to manufacturers reduce the average margin because of their far lower price, the margin will probably be even higher on retail licenses which cost far more) profit margin of 86% on Windows here, Microsoft would need to sell it at € 22 to break even (135 ÷ 157 × 100 = 86%). Making only upgrade licenses available to force students like me who never bought a Windows license before punishes them for being ‘disloyal’ customers and forces them to pay a ridiculous amount of money for a retail license.

Microsoft possesses a dominant market position and in certain aspects a monopoly with Windows. I don’t need Windows to browse the web, send e-mails or to create text documents, Linux serves me fine there. However, in certain specific cases Windows is required. My father for example often logs in remotely to his office to access his e-mails, and the company he works for decided in all it‘s wisdom to require ActiveX to log in. That is not an open standard, but a closed Microsoft standard. So he can only log in with Microsoft Internet Explorer, which means he is restricted to using Windows to log in. My problem is that many games only run on Windows, for example Crysis. Crysis uses Microsoft’s DirectX software, which is also a closed Microsoft standard. Microsoft deliberately uses this strategy to create vendor lock-in.

What has been done so far to counter Microsoft’s dominant market position or monopoly? Microsoft has a long history of litigation for anti-competitive practices. Most notable are the recent fines the European Union imposed on Microsoft, € 497 million in 2004 and € 899 million ($1.35 billion) in 2008. How effective was it? The last fine constituted approximately 10% of Microsoft’s net yearly earnings. Microsoft’s net income for 2009 was $14.569 billion according to Wikipedia. I argue the fine isn’t so high at all, of course it is a lot of money, but if it’s only 10% of their net income I doubt if it’s enough of a stimulus to start playing fair. Thanks to their dominant market position or monopoly, their profit probably increased by a lot more than 10%.

Besides the financial punishment, Microsoft had to create a version of Windows XP without Windows Media Player. However, consumers weren’t interested in this Windows XP N edition and it sold only 1500 copies to OEM’s. Windows 7 is also available as an N-edition, but it doesn’t come with a lower price. Another result of the fines has been that Microsoft can no longer charge outrageous prices for selling interoperability information for their products to competitors. More recently in 2009, The European Union demanded that Microsoft would offer a web browser choice screen because the inclusion of Internet Explorer in Windows harms competition with other browsers. The browser choice screen has been implemented in Windows 7 as of March 1.

While I think it’s awesome that the European Union gives Microsoft such a beating, I don’t think this particular approach is fruitful. I, and probably many others don’t want to use an operating system without a media player included by default. Maybe I don’t want to choose a different web browser. When we buy a smartphone, we expect exactly that, a media player included by default and a default browser, without any compulsory browser choice to be made. The essential difference here is that there is no dominant player in the smartphone market, but that Microsoft is a monopolist or dominant on the PC market. It should be evident that it isn’t good Microsoft has such a position on the PC market and I’m very much in favor of the interventions of the EU. But I’m not in favor of how they intervene.

Microsoft should have the freedom to include or exclude whatever they want in their products, every company should have that freedom, monopoly or dominant market position or not. The problem is not in Microsoft’s product, but in the consumer who can’t effectively exercise his power (partly because of a lack of information) and because of certain closed standards in use by Microsoft. Instead of dictating what Microsoft should do with it’s product, I have some better ideas.

First, Microsoft should be forced to open source some of it’s closed standards, like DirectX and ActiveX for example which I mentioned earlier. If DirectX were open source, it would be much easier to port it to Linux for example, thus removing a hurdle to play Crysis and many other games on a different operating system than Windows. The EU could forbid selling PC’s with Windows included, but again this would infringe on the freedom of the consumer. The EU should demand that the process for receiving a refund for Windows be made far more easy than it is now. Instead of having to go through the difficult process to demand a refund from the manufacturer, the seller should be obliged to provide one. Besides that the EU could dictate that the conssumer needs to be informed of the price of the operating system separate from the total price of a PC, a very good idea which I read here.

In fact, I totally agree with Alberto Ruiz’ blogpost. The EU should take care to use interventions which have less restrictions on Microsoft’s product design, but instead provide consumers with more possibilities and freedom to abandon Microsoft’s products.

My bachelor’s thesis, produced with XeLaTeX

I’ve finished my bachelor’s thesis, which I’ve produced for the Research Seminar 3 course with XeLaTeX. Here is my thesis, and here is the LaTeX source. The thesis is written in Dutch, but comments in the source are in English. This is a large document with 31 pages and a lot of literary references. I think the larger the document gets, the more advantageous it is to use LaTeX or XeLaTeX. I can’t imagine the hassle to deal with tons of literary references in such a large document. I’m very happy with the layout. If the thesis had been submitted within the deadline it would have been graded with a score of 8,5 but I failed to do so. I needed another ten days to finish, so one point was subtracted. This had me banging my head against the wall for my stupidity to underestimate the necessary work and my inadequate time management. If only I had already started working on it during the Christmas vacation there wouldn’t have been a problem. I had intended to so, but during Christmas vacation I decided to do exactly what a vacation is meant for: enjoying my free time. Nevertheless, I’m very pleased with the grade, because I expected worse based on my failings with the initial version of the paper I submitted for Research Seminar 2.

I chose to use the Linux Libertine font which I mentioned earlier, because it provides small capitals, which my other favorite font Gentium does not. This time I made some modifications to the chapter style and the section style. I decreased the font size of the chapter title and the vertical space after the chapter title. I made the section titles display in small capitals and gave the section numbering a hanging indent so that they are placed to the left of the margin just like the chapter title. For both the chapter and the section titles a subtle dark red color was used, I think it improves the looks because it’s less boring than pure black.

My faculty’s guidelines dictated that the thesis had to use an increased line spacing of one and a half. I don’t like this rule because it defaces the layout of the document and because it demands more dead trees, the document needs 12 more pages compared to standard line spacing. Fortunately the Memoir package provides the {DoubleSpace} environment, which only affects the body of the text and leaves footnotes, headers and footers unharmed. I used double line spacing instead of one and a half line spacing as specified by the guidelines, because Memoir’s double line spacing looks more similar to one and a half line spacing as produced by Microsoft Word or OpenOffice.

On the 29th page you can see a nice map which I produced myself. It shows the extent of the Greek world in the Mediterranean in the sixth century BCE. I had to create it myself because comparable maps found on the Internet were ugly bitmap images available in a low resolution. Of course bitmap images loose detail as you zoom in on them, unlike vector graphics. There was no suitable vector graphics map of the Mediterranean available on the Internet, so I asked one of the map makers on Wikipedia if he could create one for me. Unfortunately I didn’t receive any reply, so I decided to do it myself. I downloaded this map from Wikipedia and started modifying it with Inkscape. At first it didn’t work out and I was tempted to give up, but I persevered and was successful in the end.

Let me describe the methodology. The entire world map is huge, not only in canvas size but also in file size, and I only needed the Mediterranean. First you need to remove useless continents like the Americas, Antarctica and Australia and all the islands – no offense meant to the humans and penguins who live there – because those are separate objects which you can easily delete. Europe, Africa and Asia are a joined land mass so they are a single object. You need to cut it up so you remove the land which is not necessary to be visible on the map of the Mediterranean. You do this by drawing a large square over the land which needs to be removed. Then you select the square and the land, and you choose Path → Division from the Inkscape’s menu. This separates the piece of land covered by the square from the land not covered by the square. Now that the unnecessary land is a separate object, you can delete it easily.

Now you should have narrowed down the world map to the Mediterranean. Here comes the labor intensive part. Say you want to designate the area in the French Riviera colonized by the Greeks with a red color as I did. You select the land mass, then you copy it. Then you choose the Edit path by nodes button. Select the nodes on the coast of the French Riviera which you need, then choose Edit → Invert Selection from the menu. Then delete all other nodes, and clean up any remaining ones not in the French Riviera. Because all nodes are on the coast line now, you need to create a few new nodes so the object will cover some distance inland. Make sure the fill and outline color of the object are both dark red to get a comparable result to my map. Repeat this process as necessary, because I needed many small objects to mark all Greek settlements. If you do it correctly, the red objects will overlap perfectly on the landmass and coastlines, and it will not overlap on the light blue sea water. The only thing which I forgot was adding a black line (a square with transparent fill but a black outline) to the edges of the map. As you can see, on my map the outline color of the landmass borders the edges of the canvas, which is the same as the color for the coast lines.

Then increase the canvas size so the map won’t look so small and require lots of zooming in. Finally, save the map in PDF format so you can use it in LaTeX. I needed a lot of time to get this map right as you can image, but I’m certainly proud of it. Hopefully this small tutorial I’ve given here is helpful for other who encounter problems with Inkscape.

Now I’d like to mention the content of my thesis. It researches differences in Ancient Greek attitudes and practice of nudity in two dimensions, time and gender. The Archaic Period and the Classical Period to be exact. Regarding female nudity I remembered the case of Phryne, on whose Wikipedia article I had somehow (don’t ask me how!) stumbled upon a long time ago. The Wikipedia article reports that the ancient Greek author Athenaeus of Naucratis wrote that she was acquitted of a capital charge by the judges because she showed her nude body to the court. In my search for an explanation for this event I searched for scholarly articles with search engines (Google Scholar really shines here). This event required explanation because female nudity in public was not done in the Ancient Greek world in general, Phryne would be an exception to the rule. I found the article ‘Hyperides and the Trial of Phryne’ by Craig Cooper. He argues that there is convincing certainty that the event was a fabrication by a biographer who was the source of Athenaeus’ information. So the event never happened in reality, Phryne would have begged to the judges for acquittal without revealing any nudity. This means that the Wikipedia article is misleading; I intend to fix this by rewriting the article partially, so that Cooper’s perspective will also be mentioned.

Finally, I still encountered a few problems with using (Xe)LaTeX. I wanted to put a literary reference in the caption of the map on page 29. That made XeLaTeX choke during the compilation of the document, giving the following error message:

! TeX capacity exceeded, sorry [parameter stack size=6000].
\MNR@old@caption #1[#2]->
                         \def \@currentlabelname {#2}\NR@@caption {#1}[{#2}]
l.291 ...de zesde eeuw \autocite[\nopp 61]{noble}}

Googling for the error message revealed to me that it had to do with memory limitations of LaTeX, and that you could try to increase it in order to solve the problem. That didn’t help me, I didn’t dare to mess with the LaTeX configuration files too much because I was afraid it might give me further problems. To create an unnumbered chapter like I did in my document for the introduction (‘inleiding’ in Dutch) you have to use \chapter*{Inleiding}. But the consequence is that the chapter title will not be added to the table of contents and the header. I used \addcontentsline{toc}{chapter}{Inleiding}\chapter*[Inleiding]{Inleiding} to get it working correctly, but that’s an ugly, hackish solution. Why the assumption that I don’t want unnumbered chapters to appear in the table of contents? Lastly, when I used \setlength{\beforesecskip}{10pt} and \setlength{\aftersecskip}{10pt} to change the vertical spacing before and after a section title, I noticed that the paragraph following the section title would be indented. Of course that’s not correct behavior because the first paragraph following a chapter or section title usually isn’t indented. Not sure whether this is a bug or intended behavior. I still need to figure out how to fix these problems correctly, but if anyone cares to comment because they have a solution, I’d appreciate it.

Postmodernist nonsense?

Today I was searching the Internet for ideas on writing a position paper for a course I’m following this period. The course could be translated as ‘fundamentals of history’ as it aims to teach students the principles of the science of history. For this course a position paper needs to written consisting of approximately 1.500 words, this seems like a walk in the park after you just finished writing a bachelor’s thesis of 12.000 words. I’ll blog about that later. The subject of the position paper I had in mind was criticizing the Sonderweg thesis in favor of the chaos theory.

While searching on ‘chaos theory hitler’ (to filter results which relate chaos theory with nazism, because the Sonderweg thesis is related to nazism) I stumbled on this blog post by Anders Rasmussen. Rasmussen writes many interesting things, but I was specifically interested by his post on postmodernism. I agree with him that certain texts can utilize a smokescreen in the form of difficult language, making it difficult to dissect the information they provide and that once dissected the provided information could disappoint. I’ve been told Foucault and Kant can be hard to understand, I haven’t read those myself. I also often question the benefit of certain areas of science, a lot of scientific knowledge in the field of the humanities doesn’t provide the public or scientists with any benefits except for the satisfaction of our curiosity.

Rasmussen targets Judith Butler specifically. I agree his citation of Butler is incomprehensible, but I’ve had to read Butler during the course ‘Research Seminar 3’ which had ‘history of the body’ as it’s subject. Butler’s insights on gender were quite interesting in my experience, fortunately Butler has also written works which are sanely understandable.

Testing the Intel X25-M 80 GB Postville’s performance on the 8371

Encouraged by a comment on my previous post about my Acer TravelMate TimeLine notebook, I have benchmarked my Intel X25-M 80 GB Postville solid state drive using IOzone. My results are as follows:

Iozone: Performance Test of File I/O
        Version $Revision: 3.308 $
        Compiled for 64 bit mode.
        Build: linux

O_DIRECT feature enabled
Auto Mode
File size set to 262144 KB
Record Size 4 KB
Record Size 64 KB
Record Size 512 KB
Command line used: iozone -I -a -s 256M -r 4k -r 64k -r 512K -i 0 -i 1 -i 2
Output is in Kbytes/sec
Time Resolution = 0.000001 seconds.
Processor cache size set to 1024 Kbytes.
Processor cache line size set to 32 bytes.
File stride size set to 17 * record size.
                                                  random  random
    KB  reclen   write rewrite    read    reread    read   write
262144       4   38211   42784    46048    45918    8684   37782
262144      64   69923   77192   115359   113808   91553   74527
262144     512   77909   60055   220997   221988  204858   77677

Through Google I found comparable benchmarks. This one was posted on the Ubuntu Forums. It should be noted that this is a 160 GB X25-M, the poster mentions that his one is a ‘G2’ which means that it’s a second generation one with the Postville code name like mine. Probably the greater amount of storage would have some benefit for performance, but I’m not sure. These numbers are taken from the benchmark without TRIM. If I understand correctly it doesn’t matter if you use an X25-M with the latest firmware which supports TRIM (like I do), because there is no support for it in Linux/Ubuntu yet and it looks like it won’t be in the next Ubuntu release either. With Google or in that topic you can find an explanation on how you could use a recent version of hdparm and some kind of trick to use TRIM, so that’s how that poster probably got his follow-up benchmarks with TRIM. I didn’t bother because I think I’d rather wait until the support for TRIM is mature enough for it to work out of the box. The poster used an HP Elitebook 8530p with an Intel Core 2 Duo T9400 CPU and 4GB DDR2-800 RAM.

                                                  random  random
    KB  reclen   write rewrite    read    reread    read   write
262144       4   55854   61601    77975    77408   18740   37199
262144      64  102575   87223   200613   201029  141870   70205
262144     512  110951   93840   244588   242498  233184   95013

I found a second post with benchmarks on another forum, but unfortunately no more than that. In the specific post I just linked it’s not mentioned, but in an earlier post in the same topic the poster gives the model number of his SSD, INTEL SSDSA2M080G2GC, which means he has the same model I have. He posted his benchmarks at 5 December mentioning that it they were made with the most recent firmware. If I recall correctly, that’s still the latest firmware at this moment. So he’s using the same firmware as I am, the first firmware to include support for TRIM. Not sure what system was exactly used for the benchmark, but the poster mentions it’s a notebook. I’ve asked him and I’m waiting for a response.

                                                  random  random
    KB  reclen   write rewrite    read    reread    read   write
262144       4   39360   46785    53897    51421   11412   39529
262144      64   71098   54363   130520   129911   98805   74485
262144     512   81657   78925   207837   210842  218651   81242

So which conclusions can be drawn from this? No definitive. I should also take into account I’m using the EXT4 filesystem on my X25-M in combination with an alpha version of Kubuntu 10.04, which uses the 2.6.32 kernel. Benchmarks done by Phoronix show that with this and other recent kernels EXT4 suffers from performance regressions. The numbers presented by the benchmarks done by the poster  on the Ubuntu Forums leave my X25-M in the dust, but comparing to the last benchmarks doesn’t give such a dramatic difference. The greatest difference can be found in the benchmark with the 4KB blocks (first row). If anyone has a better interpretation of these benchmarks and the context to offer, please comment.

Edit 12 February 2010: the following results were achieved with the ext3 file system, using the noatime option. Contrary to my expectations it’s not better, but sucks more.

                                                  random  random
    KB  reclen   write rewrite    read    reread    read   write
262144       4   45452   42928    43576    43587    9120   40782
262144      64   59020   73224   109954   108622   90028   74706
262144     512   81555   82546   172479   171934  174380   82114

WordPress refuses to allow to upload a .tex file

For my post covering the usage of XeLaTeX I wanted to upload a LaTeX source file, a .tex file, to WordPress’ Media Library. I was surprised to notice that WordPress refused to do this because of a security policy. I posted a topic on the WordPress support forum to ask for help. When that topic had shifted to the archive pages after a day I decided to consult the IRC channel to receive support. Soon I was told that I could use the PJW Mime Config plugin to allow it. All I needed to do was give the file extension and the MIME-type, application/x-tex, to fix the problem. The plugin’s page referred to a few forum topics on the subject which explain why this was done for security reasons. I can understand that, but it’s not a user-friendly solution. It was difficult and time-consuming to figure out why WordPress refused it and how I could solve it.

KDE and Linux are power hogs

Acer claims my TravelMate TimeLine 8371 notebook should deliver over eight hours of battery life. However, that was measured with Windows Vista Business as it’s operating system. I’m using Linux, and because Windows and Linux are different the power consumption should be different as well. The long battery life of 8371 was an important selling point for me, so I decided to measure how much power is consumed by various Linux distributions with PowerTOP.

The distributions I tested were Ubuntu 9.10, Kubuntu 9.10 and the Fedora 12 with KDE. I also wanted to test OpenSUSE 11.2 and Chakra Alpha 4, but while they didn’t work on my 8371. When I tried to boot from the USB flash drives containing the Live systems of those two distributions by using the boot menu you get when press F12 on the 8371, the boot menu froze. My desktop had no problem booting from the USB flash drives with OpenSUSE and Chakra, so I assume my USB flash drives are fine and that it’s Acer’s fault because there is a bug in the BIOS of the 8371. If I’m correct this freezing occured with BIOS 1.18. I haven’t tried this with the latest version, BIOS 1.27, yet. On the other hand, considering that my 8371 had no problems with the Ubuntu, Kubuntu and Fedora Live USB’s, OpenSUSE and Chakra must be doing certain things differently which causes my BIOS to choke on the Live USB’s with those two distributions. I’ve posted topics concerning these problems on the forums of the two distributions here and here. On Chakra’s forum a solution was suggested, I haven’t tried if that works yet. An alternative is to install them with PXE so the network interface (ethernet port) can be used, but I’d rather not do that because it requires more effort. Besides giving the suggested solution a try, I’ll make sure to upgrade to the latest BIOS again. I’ll send Acer an e-mail to ask if they have a changelog of their BIOS versions as well.

Of course I used a common testing methodology for a good comparison. I installed all three distributions to my Intel X-25M solid state drive. I didn’t mess with power management settings, so all settings were at their defaults. As far as I know power management settings are the same for all three, because all dim the display when the computer is idle and reduce the backlight brightness when the notebook is on battery power. I always turned Bluetooth off with the keyboard key combination, because it’s on by default. I have to figure out a way to disable it by default so I don’t have to disable it everytime I start up my notebook, because I never use Bluetooth. I tested with WiFi on and off to see what impact it would have in power consumption. PowerTOP is available in the standard repositories of all three distributions. I made sure to start PowerTOP as root (sudo on Ubuntu and Kubuntu) because it should show more information then. I waited a few minutes to see if the reported power consumption remained steady and then noted it. Besides the X25-M which was used to replace the hard disk drive, my 8371 was completely standard and nothing was connected via USB ports. Measurements were made after starting up the Linux distributions and opening Konsole or the GNOME Terminal immediately, no other applications were started. I didn’t follow any of PowerTOP’s suggestions to save more power, I was interested in how much power the distributions would consume out of the box, without any tweaking. Desktop effects were enabled on all distributions if I remember correctly. One last thing to be mentioned is that all measurements were made while the 8371 was running on battery power, if it’s not PowerTOP won’t even work as far as I know.

Power consumption in W by distribution
Distribution WiFi on WiFi off
Ubuntu 9.10 9,7 8,5
Kubuntu 9.10 11,6 10,3
Fedora 12 KDE 11,1 9,6

Now contrast this with the figure of 5,9 Watt for Microsoft Windows Vista Business 32-bit mentioned here. Adding insult to injury, the 8371 version tested there used a dual-core Intel Core 2 Duo SU9400 CPU and 4 GB RAM while my 8731 uses a single-core SU3500 and 3 GB RAM. The difference between these CPU’s is notable, because the SU9400 is rated for a thermal design power of 10 Watt while the SU3500 is rated for 5,5 Watt. That’s a difference of 2,6 Watt considering the lowest power consumption I measured with WiFi disabled on Ubuntu 9.10. A difference of almost 30%. And then we haven’t even taken into account the lower power consumption of my CPU and solid state drive!

On the Ubuntu Forums a person who possesses an Acer Aspire TimeLine 3810T (the consumer version of my TravelMate TimeLine 8371, using the same SU3500 CPU) also investigated the power consumption. According to him Ubuntu 9.10 uses 2,4 Watt extra compared to Windows Vista. Another person who replied speculates that the cause might lie in the Linux software because it’s power management is simply not as efficient as in Windows Vista.

Besides that, according to my investigation, there is a very significant difference in power consumption when KDE and GNOME are compared. The only difference between Kubuntu 9.10 and Ubuntu 9.10 is that Kubuntu uses KDE and Ubuntu uses GNOME, so KDE must cause the extra power consumption of 2 Watt. I need to figure out if anything can be done to fix this relatively high power consumption caused by Linux and KDE and what causes it. I will write another post when I know more. Meanwhile I’m disappointed that Linux is inferior when compared to Windows Vista (and by extension probably Windows 7) when it comes to power management.

Using XeLaTeX

According to the WordPress.com Stats plugin which I use on this weblog the search term ‘memoir latex’ is the third most popular search term for my weblog. That, and considering that I’ve got something new to discuss, warrants another blog post on the topic. In my previous post on the subject I gave a paper for Research Seminar 2 as an example. That paper received an inadequate grade and I had to correct it. I have done so succesfully, and while I was working on it I also decided to use XeLaTeX for the improved version of the paper. The PDF document can be found here and it’s source here.

The important change in the improved version of the paper is the use of XeTeX, which is used in LaTeX through the xelatex command. If I’m correct all you need on Ubuntu is the texlive-xetex package. In summary the advantages of XeTeX are that you can very easily use fonts already installed on the system, that you can input Unicode in the source file and the advanced typography features which should result in better looking documents. The ability to use system fonts is very useful, because the ‘old’ method of packaging a font for LaTeX is beyond my comprehension. In the Dutch language the double dot and other diacretical marks are not uncommon, so it’s very useful to be able to input them in a LaTeX document without having to use the special LaTeX ways to get them to show up. I’m not sure if my new document using XeTeX is looking much better than the old one, but I haven’t explored all the options you can give to XeTeX which might improve the result. I’ll have to think about that in the future, possibly that’s food for a new blog post. I like XeTeX and I’m using it for all my LaTeX documents from now on.

The old paper used the Bitstream Charter font, the improved version uses Gentium. Charter was okay, but I think I like Gentium slightly more. Apparently Linux Libertine seems to be a good choice too. The last two fonts can be pulled in by apt-getting the ttf-sil-gentium, ttf-sil-gentium-basic and ttf-linux-libertine on Ubuntu.

Concerning the Memoir class and other things, for my next paper I want to make a better title page, because the current one doesn’t look exciting enough. I’ll also want to use a different section style, because the current section headings are a bit too prominent for me. I’d like to see the numbers of the sections outside of the left margin, just like the chapter style. I the previous post I mentioned that I needed to make a custom citation style for the biblatex package to make it conform to the guidelines for literature references given by the faculty of History of my university. However, the person who graded my paper didn’t seem to care, because one of the stock citation styles which comes with biblatex resembles the guidelines very closely. That’s fortunate, because designing a custom citation style for biblatex is still rocket science for me.

My favorite amateur and fan films

A week ago an amateur Uruguayan filmmaker reached the news because he landed a Hollywood contract after he showed the world his short film titled ‘Panic Attack’. He created the five minute long short film with a $ 300 budget, Hollywood offered him a $ 30 million budget to create a full-blown film. If he can do this much with $ 300, what he could do with $ 30 million would be nothing short of awesome. He made a very impressive short film. But there are more great short amateur and fan made films, so I wanted to mention those on my weblog as well.

Let’s start with another amateur film, titled ‘What’s in the Box?’. This one became a YouTube hit earlier in 2009. It was created by a Dutch student who claims it was done for € 150. According to an interview he was contacted by Hollywood as well.

There are a lot of fan films which are very impressive as well. Escape from City 17 is based on the video game Half-Life 2. Star Wreck: In the Pirkinning is a great parody on Star Trek and Babylon 5. The Star Wars fan film Imperial Military Personnel Stories takes the stormtroopers as it’s subject. Note that the last two fan films are feature length, full blown films.

Scroll to Top