Your new door to the world the slogan for Bangkok s new Suvarnabhumi airport tries to get people to stop-over on their trips between Europe and Asia. The airport s webpage claims that the airport currently handles 45 million passengers per year, but that it would scale up to 110 million. I don t see this happening, at all. Today, leaving for Melbourne, I waited for an hour in the security queue, as four flights were leaving concourse E (which has 10 gates) at once: around 500 passengers were queueing to be scrutinised for explosive liquids and other stuff a real terrorist would never consider smuggling on a plane via the standard passenger processing theatre in four lines behind four security checkpoints (in fact, that s how I confirmed my estimate: around 30 seconds per passenger (I timed it), it took me one hour, that s 120 passengers per queue, so about 500 total, at once). Four lines, one hour? 45 million passengers. 110 million passengers? Six lines (there isn t much more space), so 200 passengers per line, 1:40 hours? My new door to the world? Bangkok? Of course, this is in large part because of idiot passengers. While I was waiting in line, they came three times (every 20 minutes) to empty the garbage can where potential terrorists have to dump their drinkable explosive fluids. We ve had this ban for over 1 years, for crying out loud! Sure, some people don t (get to) fly a lot and a few may really not have heard about dangerous liquids yet (I felt really sorry for an elderly couple who had to ditch pretty much all their toiletries leaving Zurich for Madrid and wondering how they would survive there), but then at least open your eyes and turn on your brain, please. Unfortunately, the typical Bangkok/Southern Thailand tourist probably hasn t got brain cycles left between walking and carrying luggage. If you don t believe me, stop-over in Bangkok and look around: anything with legs in shorts and flip-flops probably fits right in, any obscenely and annoyingly loud group as well. For instance, this circus of 10 en route to London, who joined the end of the queue (read: about 1 hour to go) after their flight s final call with millions of shopping bags. They furiously tried to raise hell, but failed and probably missed their flight. Observing all such episodes around me made the hour go by fast I also had ample time left before my gate would close. There was one situation, however, that made walk over and complain to the supervisor (the one with the walkie-talkie): a couple from Canada were similarly faced with the final call for their flight to Toronto and walked straight up to the top of the lines outside the ropes. After a short discussion, one of the security attendants escorted them to the gate without subjecting either their bags or bodies to the ever-secure screening process. The supervisor (the one with the walkie-talkie) didn t speak English. All she did was smile and make motions with her hands which suggested that I should join the end of the queue. As I said before, I don t believe that the (minimal) added security we have at airports today corresponds to or justifies in the slightest bit the shit we have to go through these days to be airborne. However, if you make rules, then stick to them and don t open gaping (= inviting) security holes. Fortunately, the Canadians later had to surrender their weapons of plane destruction since, on every gate, another party of security attendants glove-hand-searched every bag a second time, with the previous security checkpoint no 100 metres behind. I don t see Suvarnabhumi as a successful or scalable airport. If Bangkok wasn t such a great place ?to kill a few hours (especially compared to e.g. Singapore), or the (best) entrance to Southeast Asia, I d probably avoid it in the future.

Wow, I think I got more "search results" than Google from yesterday's lazyweb search on HTML formatting. <grin> It turned out to be quite simple, as over 40 people were kind enough to point out to me -- in HTML, images are textual, "inline" elements, and so they get treated like text, in that if there's whitespace (any amount) between them, the browser treats them as separate "words" and puts some space in there. That also explains why they moved depending on the font size I chose -- the size of inter-word spacing is, of course, dependent on your font size. I'd like to thank everyone who responded and helped me out. In addition to getting the answer to my problem (which was simply "remove all whitespace between the image tags and butt them right up next to each other", like this: <img ... /><img ... /><img ... />) I got a couple of other useful related tips (like switching the images to display: block and then treating them like a div with floating and such). As much as it pains me to say it, though, I think I'll be following the tip given to me by one respondant: "just use tables". Yes, the gods of good web design shall smite me, but it is insanely difficult to get a really good, cross-browser layout happening using just CSS. It was fun to play the tweaking game the first couple of times (my blog layout is all CSS, for example, and my brigade's website, which I recently re-did, is all CSS), but each of those took me hours to get what are, at the end of the day, pretty basic and not particularly good looking sites. When all is said and done, it's just too damned hard.

This weekend has been an interesting mix of activities. Mostly I've been tweaking my mail filtering service now that it has more users it is more interesting to do that. The basic process of mail-scanning is pretty simple, but there are some fun things in the mix which make it slightly more fiddly than I'd like. The basic recipe goes something like this:

• Accept mail.
• Validate the mail is addressed to a domain hosted upon the machine.
• Do the spam filtering / magic (many steps missing here)
  • If the mail should be rejected archive it to a local Maildir folder and bounce it.
  • If the mail should be accepted then forward it to the destination machine.

The archiving of all rejected messages is a big win. It means that if there is a mistake in the handling of any mail we could undo it, retraining the spam database etc. It also provides, via a web page/rss feed, a way for a user to see what a good job the filtering system is doing - by saying "Here's what you would have had ..". Today I switched the way that the archived mail is displayed via the Web GUI. Previously I used some nasty Maildir parsing code, but now I'm running IMAP upon localhost - so the viewing of messages is a lot more straightforward. (via Net::IMAP::Simple.) More interestingly, to most readers I'm sure, today I managed to take a new Kite out for flying. A cold and windy day, but lots of fun. There was beer, pies, and near-death! This was also the second weekend I carried out some painting of my front room. At this rate I'll have painted all four walls of the room in less than two months! (The last time I painted a room it took approximately six months to complete. Move furnuture & paint one wall. Wait several weeks, then repeat until all walls are complete!)

CSS is a lovely principle, but between browser bugs and an insanely complex standard in practice, it's not nearly so fun. Today's case in point: I want three images stuck right next to each other. I've got an example of the sort of thing I want to do -- in that example, I want the three coloured squares to be adjacent, not separated by several pixels of whitespace, and for there to be no white between the boxes and the bottom border. In order to make this happen, at the moment, I've got to set negative right and bottom margins, which to me is just completely screwed up -- negative margins, to me, means overlap, not adjacency. What makes it even worse is that different browsers need different pixel counts. So far I've tested on IE6 and Firefox 1.5. The thing that really does my head in, though, is that fact if I change the font size the gap changes. So somewhere inside of both IE6 and Firefox 1.5, there's some relative size getting in the way of just sticking pictures next to each other. Update: I've been given the solution. Thanks to the 50 or so people who responded; it's great to know there's so many helpful people out there.

New York is a pretty crazy city but who knew there was the room for this type of roomy spa crazinesss?. All I gotta say is check out the different, (some very high-tech), sauna rooms. I gotta check it out. Soon. Update: A small crew went and it is, I think, the best 30 bucks one can spend in the greater New York area. The saunas, baths, and wrist watch tech-keys were nothing short of great, as was eating Korean food after a series of saunas. I can’t wait to go back during the week when I am sure it will be a little more calm.

After a few amazingly efficient days at home, working on netconf (but not getting to where I wanted) and preparing the presentation I deliver on Thursday at LCA 2008, and tying up other loose ends, I found myself in front of a television set Friday morning to witness the beginning of Federer s defeat in the Australian Open half finals. Unfortunately, I had to leave at the start of the third set and no-one in the airport knew of the game, so I had to call a friend from the plane to get the final result: 7:5, 6:3, 7:6, Djokovic won. I like Federer, and enjoy the Swiss fandom from everywhere, but his tennis wasn t up to his normal level, a true shame. Djokovic has been my favourite for about a year now, so I am excited by the result, but it would have been better to see a bit more of a battle between the two. I trust he ll win on Sunday and cross my fingers! It kind of bites to arrive in Melbourne on the day of the finals, knowing that I would have had a chance to get a ticket and could have travelled a day earlier. Oh well So I am sitting here in a crap Bangkok Internet cafe, finishing off my presentations. In addition to the netconf talk, I am also talking twice at the Debian Miniconf on Monday: first, the State of the Debian project address , which I agreed to do very short notice but knew that I could count on my colleagues to help (don t read the thread before the talk ); second, I also have a slot to talk about version control systems and Debian packaging, a topic that has been on my mind for almost three years now, resulting in elaborate workflows and many discussions. It s also becoming more and more of a hot topic for the entire project. I am tired. We landed at 5:30 and I was shattered, not having been able to shut an eye throughout the ten hours of flight. Fortunately, they let me into the lounge in the arrival area, where I crashed in one of their slumber rooms and got close to four hours of sleep, successfully ignoring the chatter from outside why did Thai Airways lay out their lounge in such a way that the slumber rooms are right next to the staff rooms with the squeaky doors? One thing about living in Asia, where space is scarce, is that people here never seem to mind about noise and sleep through anything. So I did too. I took a shower, ate some breakfast and headed into the city. I was impressed by the new Bangkok airport Suvarnabhumi , which definitely is a step up from the old, horrific one. However, I tend to doubt their claim that they could scale from the current 45 million passengers per year to 110 million. The walkways are too narrow, there are way too few toilets around, and immigration is painfully slow despite the large number of open alleys. Now in downtown Bangkok, I am biding my time. I have been here a few times before and have a number of routine stops, such as visiting the retail manager of the Jim Thompson Company, a loveable lady who speaks fluent English and is very fun to talk to. I met her during my last trip to Southeast Asia, but have since not been in touch. I now plan to head to a massage place, where they deliver merciless (and thus excellent) Thai massages, and then on to some market or Chinatown for food. I ve been training myself on chili ( Thai chocolate ) consumption for this moment: to walk up to a food stall, order a curry, and answer Thai hot when they ask for the degree of spicyness, given that I m a foreigner. At midnight, my plane leaves for Melbourne to arrive on Sunday afternoon. I have little clue as to what awaits me. I ve been to quite a number of conferences, but always perceived LCA to be somewhat special. I am definitely looking forward!

For those unfamiliar with the project, etch-and-a-half has the goal of making the existing etch release work on more hardware, but with minimal risk to existing etch users. This will be done by *adding* new kernel packages to the existing release. Kernel selection isn't final yet - the kernel team would like to use 2.6.24 if it proves itself stable in time; 2.6.22 is the backup choice. etch-and-a-half isn't restricted to kernel updates by-definition; new X drivers, or proven-safe X driver updates are reasonable candidates as well (though at this point, no changes have been proposed). An oft-asked question is around updating desktop packages, e.g. new versions of GNOME/KDE. This is beyond the scope of etch-and-a-half because it introduces a great deal more change than just additional hardware support, its wouldn't be a transparent upgrade to existing etch users, and we cannot say with a high level of confidence that it wouldn't introduce regressions for existing users. That said - a localized change to a package that adds support for new hardware and clearly doesn't break existing hardware is a valid candidate for any stable release. Such requests should be submitted to the SRM team like any other stable update request. The current plan for etch and a half is to release as part of a point release. 4.0r4 is the planned point release, since 4.0r3 is planned to have a quick turnaround to fix issues with 4.0r2. This means that pre-release bits will be made available via proposed-updates - testers are greatly appreciated. There is no plan to add support for etch-and-a-half to the etch branch of d-i. Rather, support will likely be added in a beta of the lenny installer. A wiki page exists here, and a coordination list has been setup on teams.debian.net.

I've got an upload of gpe-tetris pending - see 462306 - to provide a smaller version of the popular falling blocks game for embedded devices. One or two things crop up from this package:

1. popcon - a lot of maintainers put a lot of store by popcon results but the popularity-contest package is perl and perl is not available in Emdebian (it is thrown out along with "Essential: yes" for space reasons). Writing a replacement for popcon may not be particularly sensible either because embedded devices, although often left running for long periods in a "low power" mode, often have rare and intermittent access to any form of internet connection. This could easily result in maintainers thinking that because a package has few (or even zero) popcon figures, it is not used. On the contrary, the package could simply be used on devices that either do not support perl (Emdebian) or which do not have the ability to file popcon reports despite being able to support perl. (With the probability that GPEPhone will be added to the GPE support already in Debian, users may not wish to have popcon "phoning home" because it would likely cost real money to do so - even if a C version of popcon was written.) So I'm posting this here so that I can refer back to this post if queries arise later on.
   


2. GPE apps on Debian - gpe-tetris is inherently designed for small devices with small screens and relatively slow processors, just like the majority of GPE. gpe-tetris is also designed to be usable on a handheld device where the lack of a normal keyboard can make game play awkward. As a result, playing gpe-tetris on an average Debian box is likely to appear trivially easy and the "progression" between levels would appear very slow in comparison with the tetris available in GNOME or KDE. This is deliberate and is expressly intended to allow the game to be playable on embedded hardware that has fiddly little keys. (I'll be adding this to the manpage and a brief summary of this to debian/control.) This does give the opportunity for a little sleight-of-hand such that a device could be provided with a highscore file created on a much more powerful device and so preventing any new highscores being set, but please don't do that - play nicely!
   


3. small apps != bad code - Yes, lots of GPE packages are very small, some were clearly written originally as "demos" or during the process of learning how to write larger packages. This does not necessarily mean the packages contain bad code. I will feed back sensible patches to upstream to resolve flaws in the code but please do not think ill of the packages merely because the source code looks "trivial" compared to a GNOME package.
   


4. dpkg-shlibdeps - one issue that I must begin to track after Fosdem will be to aggregate the "foo should not be linked against" messages from dpkg-shlibdeps in GPE packages and try to build a complete picture, in case there is a real possibility of completely dropping some packages that always seem to appear in that list due to the pkg-config --libs gtk+-2.0 listing. If so, I'll begin discussions with upstream about manually specifying the GTK_LIBS variable whilst retaining the PKG_CHECK_MODULES macro that ensures that the -dev package is available in the cross-building environment. This is likely to be some form of database-query mechanism - feeding data in from grepping the build logs, building a query from the list of packages to be used for a specific Emdebian root filesystem (machine/variant specific) and then spitting out whether particular package sets can have particular targets removed. If anyone fancies working on that already, the Debian version of the information is already available via the buildd network and the Emdebian cross-build results will not differ greatly from those at this stage.
   

Other uploads include updates of libgpeschedule, gpe-calendar and gpe-clock if gpe-announce makes it through NEW as expected. Updates for emdebian-tools and apt-cross and likely new releases for QOF, gpe-expenses and pilot-qof.
debcheckout compliance
In other developments, I've worked out how to specify Vcs-Cvs correctly such that debcheckout can find the repository data correctly. Sadly, vim syntax highlighting does not help with the CVS version but the CVS syntax is:

Vcs-Browser: http://alioth.debian.org/plugins/scmcvs/cvsweb.php/dpkg-view/?cvsroot=dpkg-view

Vcs-Cvs: :pserver:anonymous@cvs.alioth.debian.org:/cvsroot/dpkg-view dpkg-view

Note the colon prefix to pserver and the space between the repository login location and the module name. Remember to omit the "-d" - debcheckout will add that for you (and won't check to see if you have already specified it).

... because they are bouncing.

In my ongoing attempt to investigate and fix problems of our mailinglist-server, i found that we only had a working bounce-detection for lists starting debian-* minus debian-private and the three digest-lists.

So i rewrote some parts of the bounce detection:

We now have four categories of lists:

• high (30 or more mails a day)
• medium (20 or more mails a week)
• low (10 or more mails a month)
• very-low (the rest)

for these four categories we calculate a spam-rate depending on the number of distributed messages and the number of bounces in the timerange of 24 hours for 'high' lists, 7 days for 'medium' lists, 30 days for 'low' lists and 90 days for 'very-low' lists.

If the Spamrate exceeds 80% for unmoderated, and 60% for moderated lists, and bounces have been seen for more than 24 hours we kick. (these rates are subject to change, when we have implemented the warning mechanism)

So these changes implemented a bounce-detection for debian-changes-digest and today we reached the level of 60% bounces for more than 500 subscribers.

So what next?

To finish the bounce-detection, i want to implement something which tries to notify subscribers about bouncing mails. This is useless for 'hard'-bouncers, but people with some smaller problems like mail2news-gatewayers this could be helpful. I also want to implement something which sends out unsubscription notices for three weeks after the kick, so kicked out people can resubscribe after they have sorted out the problems.

Both should fire once a week per address.

What do you think?

Nick Leeson: The financial storm is not over yet

Surreal to see his byline on an article the same day as a rogue trader comes to light... "Cutting interest rates has stemmed the flow for now, but 2008 will see higher energy costs, higher food costs and higher importing costs which will damage consumer confidence. Right now we can only look to Davos and hope that the global economic leaders can keep positive and stop the markets from spiralling further downwards"

SFinfo TV Schedule for Sunday

SFinfo is carrying some of the Davos meeting live again (where corporations tell governments how to direct their populations). Only interesting thing left in my opinion is the Climate Change forum on Sunday morning. Watch and see where our glorious overlords will be taking us.

How to block all Facebook application and message spam

Practical freedom from Matt Lee's Exploring Freedom.

Valuing Users by Allowing Comments : David Lee King

It makes sense - you know it does.

Internet Psychology: Podcasting? That's old-fashioned - video is the online future by Graham Jones, Internet Psychologist

Maybe I just forgot the obvious, or maybe we haven't really seen what people can do with podcasting yet.

Gaba en San Pancho Blog Archive what things to check when buying an used car

Worst used car I ever saw had a split hose leaking petrol from the carb feed onto the engine block - switching that one off when I asked was the fastest I ever saw a car salesman move!

Modern Communicator: Imarco Activ-Media have been acquired

Personally, I find one of the benefits of working in cooperatives is that it's very difficult for us to be acquired and good collaboration is much more common than merging.

sab39 ... O noes, somebody I don't like did something I agree with!

I feel the point about the bozo bit is well-made. Is one of the few features of my weak memory a low probability that I keep the bozo bit set in the long term?

How You Can Fine-Tune Your Blogger Personality Perception

How do you repair a broken perception? Can you repair it, or is it time to delete the blog and start again at a new address? Can you start again, or does web.archive mean past mistakes will hurt you forever?

robmyers - My Ogg Player

I never succeeded in accessing anything besides basic file upload/removal on my YP-U2 and Samsung Support broke promises they made to me.

another blog is possible Give cyclists a break

"Those of you drive should read this article and change your behavior towards cyclists. You may not like cyclists in the road, but they have every right to be there."

This week has mostly involved me getting my live mail filtering site up and running with a guineapig or two. This uses a custom user interface to allow users to manage the filtering settings for an entire domain:

• spam filtering.
• virus scanning.
• greylisting.
• sender/recipient whitelisting.
• DNS-based blacklists

In terms of implementation this is an SMTP proxy which is built upon the qpsmtpd framework. I've got both the user interface and a collection of plugins reading all data from an MySQL database. The practical upshot is that if you use the service you'll get less spam, and anything that has been rejected will appear in an online browsable quarantine for a period of times allowing you to view mistakes/rejected mails. Any mail you didn't want, providing you've got the spam-filtering plugin enabled for your domain, you may send back to be trained as spam. It scales nicely, doesn't appear to have lost any mail ever in real-world testing, and could be useful.

With no standard ways of plugging input methods into Linux desktops, trying to develop one used to be difficult. No more — now there is an actual standard to address this small but important part of Linux, especially for East Asian users. Last week, I was in Beijing to chair the Input Method Sub-Working Group meeting for the 13th time over the past three years. The IMSWG is part of the larger Northeast Asia OSS Promotion Forum (NEAOSS), formed by the Chinese, Japanese and Korean governments for pushing open source software. Over the past few years, every detail of how input methods should work were discussed and debated among participants from the three countries. All had strong opinions on the architecture of the system, and at times it seemed impossible for them to agree on the specification. We solved that by getting them to write software code instead — being programmers, we found they understood each other better in computer languages than in English. Now, not only do we have an agreed specification, but we are also developing a reference implementation, IMBus, thanks to the hard work of all involved, which includes James Su of SCIM fame, and Hideki Hiura, the designer of XIM and IIIMF. James gave a nice talk about IMBus at last year’s LF Desktop Architects Meeting. I said “nearly” because the overhead of being a semi-government entity means the specification won’t be published until later this year. But all technical issues have been dealt with. With hope, this specification will work its way up to ISO and published as an ISO standard in future. ISO being ISO, this won’t happen anytime soon — I’ll talk more about that later. Update: this post now also up on the Outblaze blog.

That's what happened to my /usr this morning during my daily dist-upgrade. Sigh. I probably have too many things that I installed once to get a look at and forgot to remove. Time to clean that laptop out again..:-)

Having had a number of hard drives fail over the years I use RAID whenever possible to reduce the probability of data loss caused by hardware failure. It’s unfortunate that some machines make it impractically difficult to install a second hard drive (my laptop and some small form factor desktop machines I have given to relatives). But when it’s possible I have two hard drives running software RAID-1. I use two partitions, one for /boot and one as a LVM physical volume (PV). When using RAID I make both /boot and the PV be software RAID-1 devices (the vast majority of machines that I install don’t have budget available for hardware RAID). /boot is a small partition, approximately 100M. For a machine with only one disk I make the second partition take all the available space as there is no benefit in doing otherwise (LVM allows me to create an arbitrary number of block devices out of the space at run-time). When using software RAID I often make the PV take less than the full capacity of the disks. When the disks are 40G or more I usually use less than half the capacity. For most machines that I install or run the full capacity of the disks is not required. One deficiency of Linux software RAID is that on a power failure the entire RAID device must be checked to ensure that all disks have matching data. Reading the entire contents of a pair of disks can take a large amount of time if the disks are large, and as the size of disks is increasing at a greater rate than the speed of disks this problem is getting worse. See my ZCAV benchmark results page for graphs of the contiguous IO performance of some disks I’ve owned [1]. It seems that with a 10G disk you may expect to be able to read it all in less than 1000 seconds, for a 46G disk it’ll be something greater than 1,500 seconds, and for 300G disks you are looking at something well in excess of 5,000 seconds. Almost all disks (and all IDE and SATA disks for which I’ve seen benchmark results) have the lower block numbers mapped to the outside tracks which are longer and give a higher contiguous IO speed. So by using the first half of the disk the RAID synchronisation time is reduced to less than half what it might be (in the absence of motherboard bottlenecks). When there is no need for more than about 10G of storage space there’s no benefit in making a large RAID device and wasting recovery time. While the system can operate while the RAID is synchronising the performance will be significantly lower than normal. If the usage pattern of the machine changes such that it needs more space it’s easy to create new partitions, make a new software RAID, and then add it to the LVM volume group (I’ve done this before). So the down-side to this is minimal. When creating LVM logical volumes (LVs) I create volumes for the root filesystem and swap space when doing the install. This should result in the swap space being near enough to the start of the disk to get reasonable performance (but I haven’t verified this). I make the root filesystem reasonably large (EG 6G) as disk space is plentiful and cheap nowadays and the root filesystem is the only one that you can’t easily extend when running LVM (trying to do so deadlocks the disk IO). After the base install is done I create other LVs.

• [1] http://www.coker.com.au/bonnie++/zcav/results.html

Share This

It seems to me that the Asus EeePC (a $AU499 ultra-light laptop with only flash storage) would make a decent router. Often full desktop PCs are used as routers because they run the most common software and have standard interfaces. There have been dedicated router devices with flash for a long time, but without the ability to connect a standard monitor and keyboard they were always more difficult to manage than general purpose PCs. Also dedicated routers have limited RAM and storage which often does not permit running a standard OS. According to the best review I could find [1] EeePC has a minimum of 256M of RAM, a Celeron-M CPU (32bit Intel), and a minimum of 2G of flash storage. This hardware is more than adequate to run most server software that you might want to run (my current router/firewall/server has 192M of RAM). It could run a web server, a mail server, or any other general server stuff. It comes pre-loaded with a modified version of Debian so you get all the Debian software (which incidentally means more pre-packaged software than is available for any other distribution). Bigger versions are common, I believe that the $AU499 version has 512M of RAM and 4G of flash - I’m not sure that I could even obtain a lesser EeePC in Australia. The up-side of flash is that it doesn’t take much power as having a low power device in whatever confined space ends up housing your router is a good thing (the EeePC is listed as using less than 20W no matter what the load and idling at as little as 14W) and that it doesn’t tend to break when dropped or have any moving parts to wear out. The down-side of flash is that a sufficient number of writes will destroy it. Obviously swap is not a suitable use for a flash storage device. But a small mail server (suitable for the needs of a home or a small office) should be fine with it. Squid is commonly run on router devices, to run it on an EeePC I would be inclined to buy 1G USB flash device for the cache, then if Squid’s use destroyed the flash storage it would be easy to spend $20 and buy another. The EeePC has three USB ports and a built-in Ethernet port. I believe that the minimum number of Ethernet ports for a firewall is three, this means either one for the uplink, one for the DMZ, and one for your LAN, or two for uplinks (with a redundant uplink) and one for the LAN. The three USB ports allow using two USB Ethernet devices to provide the minimum three Ethernet ports and one for USB flash storage. One notable advantage of using a laptop as a server is the built-in UPS (the laptop battery). Many people have put old laptops into service as servers for this reason, but usually an old battery gives no more than about 30 minutes of power, while a new EeePC should be able to last for more than 3 hours without mains power. Using a second-hand laptop as a server is usually not viable in a corporate environment as laptops are all different. Repairing an old desktop PC is easy, repairing an old laptop is unreasonably difficult and often more expensive than replacing it. The low price of the EeePC makes it easily affordable (cheaper than some of the desktop machines that might be used for such a purpose) and the fact that it is purchased new at such a price means that you get warranty support etc. It seems to me that a significant disadvantage of using an EeePC (or anything other than a real server) for a server task is that it lacks ECC RAM [2], but as I’m not aware of any device sold for the purpose of running as a low-end router or sold at a price-point such that it could be used as a low-end router which has ECC RAM I guess that’s not a big disadvantage for that purpose. But the lack of ECC RAM and the lack of useful support for RAID does make it a little less suitable for use as a mail server. Finally if you are going to run a VPN then it would be handy to be able to secure the hardware against malicious activity when the office is closed. An EeePC will fit nicely in most safes… Please let me know what you think about this idea. I’ll probably start installing some such machines for clients in the near future if ASUS can keep up with the demand.

• [1] http://www.hothardware.com/articles/Asus_Eee_PC_Full_Retail_Review_Showcase/?page=1


• [2] http://etbe.coker.com.au/2007/07/15/desktop-machines-and-ecc-ram/

Share This

I have just uploaded javatools 0.5 to unstable and it is currently waiting in NEW. Anyone wishing to try it out earlier can find source and packages (it's arch: all) at http://www.srcf.ucam.org/~mjj29/javatools/ . Javatools replaces the existing jarwrapper package and also contains programs to help packagers in creating packages for Java programs and libraries. Packaging tools The javahelper package consists of several small programs which make packaging Java programs and libraries easier. They are generally designed to work in the same fashion as the debhelper programs, but start with the jh_ prefix. All of the programs have their command line arguments documented in manpages. jh_build Many Java programs and libraries are distributed without sane build systems. jh_build provides a simple interface for building Java source code into Jars, including setting the appropriate entries in the manifest. In almost all cases all that needs to be done to call jh_build is to set JAVA_HOME and CLASSPATH and then call jh_build with the name of the jar and the directory containing the source.

JAVA_HOME=/usr/lib/jvm/java-gcj
CLASSPATH=/usr/share/java/esd.jar:/usr/share/java/jsch.jar
jh_build weirdx.jar src
   

This command will compile all the Java files under src , set the classpath in the manifest and build it all into weirdx.jar. A couple of other options are worth mentioning. If this jar contains an application rather than a library then the -m or --main option can be used to set the Main-Class attribute in the manifest which will allow the resulting jar file to be be executed jh_build also provides a --clean parameter which should be called in the clean target of debian/rules . jh_libs For library packages Debian Java policy currently requires that the libraries be installed to /usr/share/java in a versioned format and with an unversioned symlink. jh_libs will take a jar and correctly install it. As with debhelper programs, this can either take a jar as a parameter, or read a list of jars from a file in the Debian directory. It also follows the -p , -i and -s semantics of debhelper for selecting which packages to install the jar to. When operating on a package, jh_libs will read the list of library jars from debian/package.jlibs or debian/jlibs . The jlibs file is a list of jars to install, one per line, and works exactly the same as listing them on the command line. Each jar is installed to debian/package/usr/share/java/ in the appropriate versioned and unversioned forms. jh_depends jh_depends works like dpkg-shlibdeps, but for jar files. For each jar in the package it takes the jars on which it depends and finds the packages to which they belong. These are included in the debhelper substvars as $ java:Depends . The control file can then just list that variable which is filled in automatically. This is done by reading the Class-Path attribute from the manifest of each jar. Jar files should include this attribute to prevent applications which use them from needing a full recursive classpath in their startup scripts and to prevent unneccessary transitions when the library changes its dependencies. If the package is not built with jh_build and the upstream build system doesn't set it correctly then jh_manifest can be used to fix this. If the application uses executable jars (see Runtime support below) then jh_depends will also add the appropriate depends on jarwrapper and the correct Java runtime. jh_manifest Many upstream build systems do not set the Class-Path attribute in the jars they create. This leads to several unwanted problems, such as expanding the classpath which applications have to use and introducing unneccessary transitions. They also may not set the Main-Class attribute. Both of these are required for running jars with the -jar parameter. jh_manifest can fix the manifest files of jars. It can either read from a manifest file in the Debian directory or run in a mode which updates all the jars with the CLASSPATH environment variable. The manifest files can either be debian/package.manifest or debian/manifest . The format of this file is a list of jars and indented below each one a list of manifest elements to set:

usr/share/weirdx/weirdx.jar:
 Main-Class: com.jcraft.weirdx.WeirdX
 Debian-Java-Home: /usr/lib/jvm/java-gcj
   

jh_exec The Runtime support section below describes running executable jars directly. jh_exec will scan package directories for jars in the paths, or symlinks to jar from the paths, and ensure that they have been set executable if necessary. jh_makepkg jh_makepkg will create template Debian packages for Java programs and libraries similar to dh-make . It should be run in the source directory and it will create the orig.tar.gz and most of the files in the Debian directory, which need only small changes neccessary to build the package. java-propose-classpath Some upstreams have complicated classpaths which may not be obvious to the packager when using jh_manifest to set the Class-Path attribute. java-propose-classpath will unpack a jar and look at the symbols imported to the class files, then scan all the jars in /usr/share/java . This shouldn't be run in the build since it is slow, and there may be ambiguities that the packager must resolve. It is still very useful for the packager as most of the time it will get it right automatically. Runtime support Javatools also provides some runtime support. Unlike compiled programs, or purely interpreted programs with hash-bang lines, Java programs cannot be directly executed. Many upstreams expect them to be run using java -jar jarname or java classname . This is not generally acceptible in systems which expect to just be able to run the command or launch it from a menu. As a result, many packagers are writing wrapper scripts which just call java with the correct classpath, jar and main class. jarwrapper There is an alternative to wrapper scripts, however. The binfmt_misc kernel module allows the kernel to call out to a program in userspace to execute specific types of file. jarwrapper registers itself as a handler for executable jars. This is done by reading values from the manifest file. In order for executable jars to work the following attributes must or may be defined in the manifest. These attributes can be set using jh_build and jh_manifest .

• Main-Class: The name of the class to be run when the application starts. (REQUIRED)
• Class-Path: The path to all the jar files on which this jar depends. (REQUIRED unless empty)
• Debian-Java-Home: A Debian-specific property if this application depends on a specific runtime. Specify the path to the runtime which should be used. (OPTIONAL)
• Debian-Java-Parameters: A Debian-specific property if this application needs extra options to the JVM. (OPTIONAL)

Putting it together This section shows the debian packaging generated by jh_makepkg for an application and a library using jh_build . Sample Library Packaging debian/control

Source: jsch
Section: libs
Priority: optional
Maintainer: Matthew Johnson 
Build-Depends: debhelper (>= 5), javahelper, java-gcj-compat-dev, libzlib-java
Standards-Version: 3.7.3
Homepage: http://www.jcraft.com/jsch/
Package: libjsch-java
Architecture: all
Depends: $ java:Depends , $ misc:Depends 
Description: Java secure channel
 JSch is a pure Java implementation of SSH2. JSch allows you to
 connect to an sshd server and use port forwarding, X11 forwarding,
 file transfer, etc., and you can integrate its functionality
 into your own Java programs. JSch is licensed under a BSD style
 license.
   

debian/rules

#!/usr/bin/make -f
export JAVA_HOME=/usr/lib/jvm/java-gcj
export CLASSPATH=/usr/share/java/zlib.jar
build: build-stamp
build-stamp:
	dh_testdir
	jh_build jsch.jar src
	touch $@
clean:
	dh_testdir
	dh_testroot
	jh_build --clean
	dh_clean
	rm -f build-stamp jsch.jar
install: build
	dh_testdir
	dh_testroot
	dh_clean -k
	dh_installdirs
binary-arch: build install
	# Java packages are arch: all, nothing to do here
binary-indep: build install
	# Create the package here
	dh_testdir
	dh_testroot
	dh_clean -k
	dh_install -i
	dh_installdocs -i
	dh_installchangelogs -i
	dh_javadoc -i --sourcedir=src 
	jh_libs -i
	jh_depends -i
	dh_compress -i
	dh_fixperms -i
	dh_installdeb -i
	dh_gencontrol -i
	dh_md5sums -i
	dh_builddeb -i
binary: binary-indep binary-arch
.PHONY: build clean binary-indep binary-arch binary install
   

debian/libjsch-java.jlibs

jsch.jar
   

Sample Application Packaging debian/control

Source: salliere
Section: misc
Priority: optional
Maintainer: Matthew Johnson 
Build-Depends: debhelper (>= 5), java-gcj-compat-dev, fastjar, 
               libmatthew-debug-java, libcsv-java,
               libitext-java, javahelper
Standards-Version: 3.7.3
Package: salliere
Architecture: all
Depends: $ java:Depends , $ misc:Depends 
Description: Short Description
 Long Description
   

debian/rules

#!/usr/bin/make -f
export JAVA_HOME=/usr/lib/jvm/java-gcj
export CLASSPATH=/usr/share/java/csv.jar:/usr/share/java/debug-disable.jar:/usr/share/java/itext.jar
build: build-stamp
build-stamp:
   dh_testdir
   # Build the package
   jh_build salliere.jar src
   touch $@
clean:
   dh_testdir
   dh_testroot
   jh_build --clean
   dh_clean
   rm -f build-stamp salliere.jar
install: build
   dh_testdir
   dh_testroot
   dh_clean -k
   dh_installdirs
binary-arch: build install
   # Java packages are arch: all, nothing to do here
binary-indep: build install
   # Create the package here
   dh_testdir
   dh_testroot
   dh_clean -k
   dh_install -i
   dh_installdocs -i
   dh_installchangelogs -i
   jh_manifest -i
   dh_link -i
   jh_exec -i
   jh_depends -i
   dh_compress -i
   dh_fixperms -i
   dh_installdeb -i
   dh_gencontrol -i
   dh_md5sums -i
   dh_builddeb -i
binary: binary-indep binary-arch
.PHONY: build clean binary-indep binary-arch binary install
   

debian/salliere.install

salliere.jar usr/share/salliere
   

debian/salliere.links

usr/share/salliere/salliere.jar usr/bin
   

In just over a month (1 month and 1 day, to be precise), it's FOSDEM again. The Debian Developers Room schedule is almost complete, and some interesting things are being published around it. Such as the pre-FOSDEM beer event. In previous years, this always happened at the "Le Roi d'Espagne", a bar at Grand Place in Brussels. This was a nice place for many years, because it's easily found, and they have this backroom where quite a lot of people would fit in. Unfortunately, "quite a lot" does not equal "a huge number", with the amount of people going to the pre-FOSDEM beer event more and more approaching the latter than the former as the years go by. Last year, it finally became too crowded, and it was decided that the pre-FOSDEM beer event would change bars this year. Which has now happened. This year, the pre-FOSDEM beer event will be at the Delirium Caf , which is not at the Grand Place anymore, but just 100m away from it; the FOSDEM website has more information. A nice feature of the Delirium Caf , besides the fact that they are much larger, is the huge amount of different beers they have. Over 2000 of them, which you can query online!. And that's not even considering the soft drinks or waters or whatever, which they also have on their menu. Another novelty this year is the streaming; the main tracks that will be held in Janson will be streamed out of the event, and the Debian video team is also planning to do some streaming. Meaning, even if you can't make it to FOSDEM, you'll still be able to follow some of it. You won't be able to follow everything, especially not the hallway track which at FOSDEM is a very important one; but at least you'll be able to sample some bits, which is nice. Other than that, not much will change. There'll be a key signing party, organized by yours truly. There'll be Lightning talks, there'll be the Main track, and there'll be a number of Developer rooms, including the Debian one—which, occasionally, is also organized by yours truly. This conference has always been my favourite (even though I haven't been to many conferences besides FOSDEM); I've never missed even one edition ever since the very first one back in 2001 (when it was still called OSDEM). And since a few years, I've also been offering crashing space at our office, for international geeks who don't mind roughing it a bit (it's an office, there no shower, there's only floor space). This year, perhaps I haven't been spamming enough for that, but for now I only have two people who're interested in sleeping over (and they're not even sure); if you're going to FOSDEM and still need a place to sleep, you're (probably ;-) welcome. Send me an email. On the Debian front, things are going well, for the most part. The list of speakers for the Devroom is nearing completion, leaving me with little more to do than to start assigning times. We've been assigned a booth as well, but there I still need to figure out how to organize it, and what to do with it. There will of course be the regular UK T-shirts, and probably some other merchandizing, but I also want something involving some computer device. We've had this d-i "babelbox" installation running a few times now, which was nice; but I do think we need some variation, too. The only problem with that approach is that I don't really know what that variation could be. Suggestions are welcome! And so time goes on, and FOSDEM draws nearer. At some point it'll eventually happen, and then it will be over before one can say "blueberry pie". Sometimes I wonder whether all these months of preparation are actually worth it. But when FOSDEM is actually here, I'm sure I'll find that it is, indeed, worth it...

For the time being offline I've been doing a lot of things, and some of them involve free software of course. I'm glad to announce that this march Peru will be heading the direction from its current status-quo to, what we expect, is the first step into a real and sustainable ecosystem where our talented people (from academy, and others) get their hands dirty into to real contribution to free software projects. involucrate.org has born as a project focused on building an environment to foster a real free software community focused on technical contribution and making our country have a larger spot on the free software contributors world-map. The lack of appropiate models and the inhability to stablish a link between the real actors on such scenary have led to a poor involvement and groups atomization for "ideological" or political means, making room for "religionization" of a very technical and licensing model matter where debate over technical aspects have gone 0. Our manifesto (in Spanish) explains our vision and mission, and also how we expect to achieve such a challenging goal. Is in our will to link our academic people, who was somehow behind the walls (a luxury for a country like mine), and the other actors here and globally. We want our talented people to see globally, and offer our life-experience on the field to make them aware and show them the opportunities that they can have and how other people near us who are making it right!. This time, thanks to the wonderful guys at GNOME Foundation, we are having our very first international conference called involucrate+GOME focused on free software development and, as we expect to be, be the first step of a series of world-class events where we expect to set a yearly world-known free software event here in Peru. Something we dodn't have after more than 10 years since the first group of people gathered together to discuss which is the best distro or something like that and the discussion is still ongoing. As noted the conference will be focused mostly, well almost 90%, in GNOME, but we expect to have more editions dedicated to one or more projects. So, I invite you to check our program and our cool speakers, all from south america and all contributing on GNOME. Of course you are invited to attend to, if you happen to like us! Bring your questions and contributions to this initiative to both our project web or DebianPeru web who is coordinating this initiative, I mean this is almost a debianite sub-project ;) I want to say as a side-note, that despite what media has made people think, we are making a real cool relationship with the chilean community. Expect to hear from us soon, and not because we've buying more wine from Chile and they are importing real pisco from Peru, or because we'll mimic our presidents. I really like this time, it comes 10 years late, however!

A conversation at lunch today inspired me to work on a quick estimate of the amount of power wasted on animation and Flash-laden MySpace profiles. My conservative estimate is somewhere in the region of 320 kiloWatts, which is much less than I'd have expected. This is probably skewed by me carrying out the test on a laptop - I'd guess that it'll be worse on desktops, but I need to do some more testing to be sure (ideally involving a Watts-up and a serial connection so it can be done without a ridiculous amount of pain). By changing a couple of assumptions it's quite easy to get it up into the megaWatt range, but we're still talking "That's quite a pile of money" rather than "That's a sizable proportion of the US electricity output".

Interestingly, the vast majority of the extra CPU usage (and therefore electricity) seems to be down to Flash. Improving its performance would be a pretty big step in reducing overall power consumption, especially given the number of websites running flash adverts. Want to save the world? Turning off Flash might be a surprisingly good step.

I've written up my methodology and results here - I'd be interested in any feedback.

Update: Why don't I learn my lesson and stop working too late? Making a fool of myself in public isn't something to be proud of.

Ignore this post if you're seeing it for the first time.

---

Martin, David, this script is broken. When trying to set kernel parameters, there should be a new line, it shouldn't be explicitly removed. I know it looks weird, but this seems to be proven by this output:

kinder:/etc/init.d# ./params4flash start
kinder:/etc/init.d# cat params4flash
#! /bin/sh

exit 0

PATH=/sbin:/bin

# Max time to wait for writeout
MAX_AGE=120
CENT_AGE=$((100 * $MAX_AGE))
# Max percent of mem to use for dirty pages
DRATIO=10
# Once we write, do so until this many percent of mem is still in use
DBRATIO=1

case "$1" in
start)
   echo -n 0         >> /proc/sys/vm/swappiness
   echo -n $MAX_AGE  >> /proc/sys/vm/laptop_mode
   echo -n $CENT_AGE >> /proc/sys/vm/dirty_writeback_centisecs
   echo -n $CENT_AGE >> /proc/sys/vm/dirty_expire_centisecs
   echo -n $DRATIO   >> /proc/sys/vm/dirty_ratio
   echo -n $DBRATIO  >> /proc/sys/vm/dirty_background_ratio
   ;;
restart reload force-reload)
   echo "Error: argument '$1' not supported" >&2
   exit 3
   ;;
stop)
   # No-op
   ;;
*)
   echo "Usage: $0 start stop" >&2
   exit 3
   ;;
esac
exit 0
kinder:/etc/init.d# cat /proc/sys/vm/swappiness
60