What's the weather like in Debian? One of the cool things developed during EDOS (now Mancoosi) has been edos-debcheck. Using it it is possible to check (incredibly quickly) whether some package in a given distribution cannot be installed according to its dependencies. When there are such packages it usually means that the distro is buggy s it is shipping uninstallable packages (there are some corner cases, but they are rare). In fact, edos-debcheck has been run daily (as a service maintained by Ralf) on Debian's main suites and some derivatives for years now. The resulting list of uninstallable packages is linked from http://edos.debian.net and is being used by the QA team; I've no idea if it is being used by the Release Managers too, but it probably should be. A cool toy built on top of edos-debcheck was the Debian weather (part of a more generic service called Anla, I'll blog about it sooner or later ...). Debian weather basically used to present the status of a given Debian-based distro in term of how many uninstallable packages were shipped using the weather metaphor: the fewer the uninstallable packages, the better the weather. I wrote "used to" because Anla was stopped about 1 year ago, and is being re-engineered these days. In the meantime it was a pity not to have the Debian weather (as Enrico pointed out with me), so I spent some time to resurrect it on top of edos-debcheck daily runs: Debian weather is back, enjoy!

Ana started to spread the world with current discussions about which KDE in Lenny. Let’s relay on Debian/KDE universe: you can follow the discussion here and participate if you care about Debian STABLE and KDE. It is a hard decision and I gave my opinion. I think it is a good compromise between stability and bleeding edge stuff.

I just got back from Indiana Jones and the Kingdom of the Crystal Skull, a movie in which Steven Spielberg takes his obsession with the extra terrestrial to the next level.

Please watch it if you feel that times goes too fast, the two hours that you are watching the movie will seem like two years.

And it had to be a UFO in the end :)

Update: I should have kept the spoiler under lj-cut, but now that I think of it, that wouldn't help people aggregating the posts. I suppose I will avoid posting spoilers.

There is a bit of discussion on Planet Debian on dbus and HAL and how to handle the daemon restarting. Basically, our "hardware abstraction" layer is actually more like a policy layer: the kernel provides the real hardware abstraction, but its access control features are not fine grained enough, and there is no revocation of privileges. Right now, there are two systems in use: X11 for keyboard/video/mouse, and dbus/hal for the rest. X11 has no problems with access control, because any process that wants to talk to the X server is owned by the user sitting at that terminal, and when that user logged off, the X server got uninteresting for us. For other hardware, that model doesn't entirely work. KVM is unique in that it only affects one user, and there is a clear definition of that user going away. To summarize my desired policies:

Sound

is similar, but I often log out in the evening and have something playing while I fall asleep, so the policy on the sound device is that access is prolonged even after logging out.

USB drives

should be accessible to the user that plugged them in, and stay with them, even after they logged out, until they are disconnected. I'd like to have block level access to them and run my file system drivers in user space (so I can format a pen drive without administrative access to the machine). That does not apply to all USB drives, though, so I'd like to be able to "recognize" certain volumes and handle them differently, not so much as a security but as a safety measure.

Smart cards

should be like KVM, really. No one but me should ever be able to talk to my cards, because causing a DoS is just so easy — enter the wrong code three times, and I have a big problem. Obviously, that is part of the security that my card provides (if you want to upload Debian packages in my name, you have to steal my wallet and get my PIN right in the first three attempts).

USB phone handset

I'd like that phone to be able to receive calls anytime, even when no one is logged in, and the currently logged on user should be able to make a call (but not over my account).

Specialty hardware

like the JTAG adapter hanging off USB on my box need a different policy too.

Any of these (except for the smart card) might mean that the device could be connected and used for a very long time, longer than I can sit at the console. Unlike humans, who get only mildly annoyed if the display closes and applications need to be restarted, there are serious data consistency issues if connections break and need to be rebuilt — and the policy needs to allow for rebuilding the connection securely. If we want to allow for handing over a resource to a new instance of the server or the client, then the protocol needs to be aware of this, as we need to synchronize server and client state after a connection breakdown. It might be possible to create a session management layer for certain services (analogous to X11 toolkits that transparently handle rebuilding the widgets on restart without bothering the application about it), but it'd probably be difficult to do without a full protocol redesign, as data loss is simply not as acceptable as when talking to a human who can pick up the pieces and go from there.

I recently noticed that a very good way to safely read webcomics in the bath tub is an old laptop with a big screen (e.g. a IBM ThinkPad A-series like my 15” A31 which has a nice 1400×1050 resolution), a water proof keyboard, the screen-alike, keyboard only driven (hence the name) window manager ratpoison (other keyboard driven window managers like wmii or awesome probably will do as well as ratpoison) and a good keyboard driven web browser which can bind or by default has bound a key to follow <link rel="next" ... /> tags. Like Opera. Opera has bound the space bar to scroll one page down and if you reach the bottom of the page to go to the next page as labeled in the link tag. Additionally the full screen mode is helpful, too. Or the dream browser of all Emacs addicts, Conkeror, which has bound the function browser-follow-next to ]]. (Conkeror packages will hit Debian Experimental quite soon.) Or the GNOME feed reader Liferea which has bound Ctrl-Space by default to scroll down the content by one page and if you reach the bottom of the content go to the next unread item. With that equipment I can read my favourite web comics like Questionable Content (whose content seldomly is questionable :-) or Ozy and Millie (Think of a mixture of Calvin & Hobbes, Peanuts and Kevin & Kell) in the bath tub without drying my hands before reading the next comic or fearing water or health damage by the combination of water and computer. I just press one or two keys on the keyboard floating over my lap and have a good time. BTW: I’ve got a blue, non-branded one (packaging reveals it as “AirTouch Keyboard”, probably manufactured by SanChuan Electronics, China) with swiss-german layout from ARP Datacom (whose website offers no permanent links and insists on session cookies *puke*), but those from Keysonic or from ROCK seem to be very similar — nowadays they are also available in illuminated, miscellaneous colors and wireless, but only IP65, probably because of the necessarily accessible battery compartment. But this kind of having fun still has optimisation potential: non-flexibel water-proof keyboard (IP67 recommended, so those IP66 keyboards and mice recently posted at UF LOTD are probably not tight enough), flat screen mounted above the bath tub, etc. ;-) Or maybe a completely water proof laptop if such thing exists — Does it? One more note: In Debian Sid and Lenny recently a new tool called keynav has been added, which allows you to control the mouse quickly using the keyboard only. So with Sid or Lenny, I don’t even need an waterproof mouse or trackball if an application insists on mouse usage. ;-)

So, the last lazyweb question I asked had good results. Time to try again: Whats a good python-accessible, cross-platform-and-trivially-installable(windows users) flexible (we have plain text, structured data, etc and a back-end storage area which is only accessible via the bzr VFS in the general case), fast (upwards of 10^6 documents ), text index system? pylucene fails the trivially installable test (apt-cache search lucence -> no python bindings), and the bindings are reputed to be SWIG:(, xapian might be a candidate, though I have a suspicion that SWIG is there as well from the reading I have done so far, and - we'll have to implement our own BackEndManager subclass back into python. That might be tricky - my experience with python bindings is folk tend to think of trivial consumers only, not of python providing core parts of the system :(. So I'm hoping there is a Better Answer just lurking out there... Updates: sphinx looks possible, but about the same as xapian - it will need a custom storage backend. google desktop is out (apart from anything else, there is no way to change the location documents are stored, nor any indication of a python api to control what is indexed). It looks like I need to be considerably more clear :). I'm looking for something to index historical bzr content, such that indices can be reused in a broad manner(e.g. index a branch on your webserver), are specific to a branch/repository (so you don't get hits for e.g. the working tree of a branch), with a programmatic API (so that the bzr client can manage all of this), with no requirement for a daemon (low barrier to entry/non-admin users).

Lessons learned: no unicode in filenames on my blog. I had the great idea, a few days ago, to write a short blog post about my trip to Cologne, and to give the file a name with a UTF-8 character in it (the , to be exact). Except that I forgot how my blog actually works... I use blosxom to manage my RSS feed and other parts of the blog; but to integrate it properly in my website, I wrote a whole lot of stuff around that. It's not just blosxom:

• All files are in a subversion repository.
• On the server, there is a file:/// checkout of the subversion repository in /var/local/blosxom. Files in this checkout are owned by the www-data user.
• The subversion repository has a rather long post-commit hook:

wouter@samba:/var/lib/svn/blog/hooks$ grep -v '^#' post-commit   wc -l
20

It takes care of updating the repository in /var/local/blosxom, and then fixes the timestamps on those files based on the timestamp of the original commit in the subversion repository. It's an ugly amalgam of svnlook date, svnlook history, touch and things, but it works. Sortof. Except if I remove a file, or if the svnlook history thing doesn't find the file itself. Finally, it will call blosxom itself, in the mode in which it creates files on disk rather than trying to do CGI output.
• In blosxom, then, I added two extra flavours: one I call myrss, and another I call php. The myrss flavour is used because I don't like blosxom's default RSS output; it doesn't do anything strange. The php flavour is what you see when you go to my blog on my website: it helps me integrate the blog in the rest of the "design" (ahem) of my website, allows me to add google ads in between my blog posts, and, finally, to hook in my home-brewn comment system (which uses postgres to store comments).

Apparently, however, subversion thinks differently about UTF-8 filenames if you do a checkout from a repository when using a http:// or a file:// URL. As a result, there were some issues in the post-commit subversion hook that I wrote, resulting in the svn up part of the post-commit not entirely succeeding. Or some such. Then, the touch is done, and the svn up doesn't work at all, anymore. In short, things started to break horribly, resulting in empty posts (because the files were created by touch rather than svn up), the comment thing being confused about filenames (and, as a result, postgres complaining about incorrect UTF-8 encodings), and similar other ugliness. So, I just renamed the file, and did a cleanup of the subversion checkout in /var/local/blosxom. I'll just have to cope with the fact that my setup doesn't like unicode, I guess. Or, perhaps, finally switch to ikiwiki some day, which I've been thinking about ever since Joey first blogged about it a few years ago. But that's not urgent...

Four years ago, the last time I was in Brazil, I came as an anthropologist-in-training to attend and give a talk at Debconf4 held in Porto Alegre. I have returned to this city in the south, again in the winter, but this time I have come back as an anthropologist who is giving a talk at the department of anthropology at the UFRGS, based on my research conducted many years ago. This time, thanks to my (really friendly and energetic) hosts, I am seeing far more of the actual city, its beautiful and legendary sunsets, its outlying neighborhoods, and even its subcultures. This has been my first foreign trip where I am interacting primarily with anthropologists and this has been really interesting for me. On the one hand, because of the language difference, there is a lot that is strange and hard for me to follow and I am acutely aware of the general and particular economic gulf between north and south and its impact on students (books are expensive, traveling for conferences is very difficult, and even applying to graduate schools in the north can be impossible because it costs so much to apply to each school!) On the other hand, the methods and subjects of study, the style of analysis, and the teaching are all very familiar, making me feel like I am part of a discipline that (thankfully) transcends national borders. Many of the projects I learned about–the tensions between free software advocacy and development, the role of memory among fan s of a corny country/Gaucho singer from this area and the surgical management of intersex (hermaphrodite) newborns were some of the most interesting projects I have come across and gobbled as much information about them as I could. I was also struck, yet again, by the deep myopias of anthropology, north and south. My own project, which focuses mostly on white, American and European hackers, often does not strike as culturally authentic enough because, well, the people I study are white, American and European (and I am slowly coming to see that if you study the so-called white and male or the North American/European elements of technology and the Internet and the law, you are probably white and male yourself and anything having to do with ethnicity/gender is usually the province of female academics, which I find really problematic). If I had carried out my first project in Brazil, where there is a foreign language, where there is a long tradition of studying various groups, then my project would have been stamped by that mysterious aura of authenticity/approval. On the other hand, two of the students here (those that worked on free software and the country singer) complained that their papers have yet to be accepted by the Brazilian Anthropological Association for being too strange, non-traditional, and it seems in their case, the problem is that they are studying urban Brazilian, popular culture (which for me would have just been just the thing to study). We are located far apart but find ourselves in similar positions by virtue of studying something geographically close to where we live, which is confused and misperceived as being culturally close to our world. But if there is something I wish to hammer into my students and other anthropologists is that there is tremendous plurality and multiplicity within our own societies. We can travel far, in the cultural sense, just by staying home and opening our eyes a little wider. Another nice experience is that I am learning a tremendous amount about free software politics and development from the graduate student, Luis Felipe, who is really responsible for getting me down here. He found me on IRC many moons ago and finally we have been able to compare notes and have long conversations about free software, the differences in how we can gather data due to gender, and a topic close to me, which is the tension between the political and apolitical in free software. Our relationship is at once based on friendship and also one of mentoring. And it reminds me of others who have also mentored me and how crucial this mentoring was to the development of my own work. One of my most important mentors, Chris Kelty, just published his book on Free Software (and I will soon write an entry about the book but the WHOLE THING IS ONLINE) and so it felt quite nice that we were discussing what I think his one of his most amazing chapters on the creation of the copyleft in our class on Friday. Not only was the topic about the genesis of the first free software license appropriate for a class on IP, but in many ways, because of his mentoring, I have gotten to where I am and so it felt also appropriate to honor and recognize that genealogy in my own work.

 158 Ns+ Jun 04 Debian Bug Trac ( 13K) Bug#250202 closed by Russ Allbery 

That only took, eh, four years. The report started out as a "I'm sick of all these confusing and different ways to package a package, we need a solution", suggesting a few extra targets to debian/rules, but then quickly changed into a proposal to add a README.source file to the debian/ directory. At some point, early on, I started to prefer the README.source solution way more than the extra debian/rules targets thing, but the proposal reached a stalemate as some people did prefer the extra targets thing. A solution was only reached when Russ Allbery wrote a compromise proposal that managed to satisfy everyone, which he first sent out about half a year ago. Now, finally, the update is in. Which makes me really, really happy.

I've released version 0.4.11.4 of the aptitude package manager (release notes). The main motivation for this release is bug #483920, which would cause the resolver to choose solutions badly (e.g., it might prefer downgrading a bunch of packages to keeping them at their current versions). It also includes a simple fix for bug #136874, the fact that the output of aptitude search is badly suited for scripts (it still is, but now there's a command-line option that will make it better).

One think that really is annoying with Launchpad is its lack of interfaces with the outside world. No SOAP interface (well, I think that work is being done on this), no easy way to export all bugs. The only way to get all the bug data in a machine-parseable is to first fetch this URL, and then, for each bug number listed there, to make another request for https://launchpad.net/bugs/$bug/+text. I filed a bug a few weeks ago, asking for a simpler way to get all the data. A Launchpad dev suggested to do what I just described (fetch all the number, then fetch the data for each bug). I originally dismissed the idea because it just sounded too dirty/aggressive/whatever, but since I needed to practice python, I gave it a try. And actually, it works: I was able to get all the data in less than an hour (but that probably put some load on Launchpad ;-)). That allows to write cool SQL queries. Bugs with the most subscribers:

select bugs.bug, title, count(*) as subscribers
from bugs, subscribers
where bugs.bug = subscribers.bug
group by bugs.bug, title
order by subscribers desc
limit 10;

bug	firefox	subscribers
188540	firefox-3.0 crashed with SIGSEGV in g_slice_alloc()	291
154697	package update-manager 1:0.81 failed to install/upgrade: ErrorMessage: SystemError in cache.commit(): E:Sub-process /tmp/tmpjP6Bsx/backports/usr/bin/dpkg returned an error code (1), E:Sub-process /tmp/tmpjP6Bsx/backports/usr/bin/dpkg returned an error code (1), E:Sub-process /tmp/tmpjP6Bsx/backports/usr/bin/dpkg returned an error code (1), E:Sub-process /tmp/tmpjP6Bsx/backports/usr/bin/dpkg returned an error code (1)	278
141613	npviewer.bin crashed with SIGSEGV	262
59695	High frequency of load/unload cycles on some hard disks may shorten lifetime	182
215005	jockey-gtk crashed with AttributeError in enables_composite()	171
216043	compiz.real crashed with SIGSEGV	168
121653	[gutsy] fglrx breaks over suspend/resume	144
1	Microsoft has a majority market share	142
145360	compiz.real crashed with SIGSEGV	134
23369	firefox(-gnome-support) should get proxy from gconf	126

Bugs where someone is subscribed twice:

select bug, subscriber_login as cnt
from subscribers
group by bug, subscriber_login
having count(*) > 1;

bug	subscriber
33065	mvo
48262	mvo
144628	skyguy
158126	benekal
213741	sandro-grundmann
216043	jotacayul
221630	kami911

(Yes, that forced me to change a primary key) Packages with the most bugs:

select package, count(distinct bug) as cnt
from tasks
group by package
order by cnt desc
limit 10;

package	number
ubuntu	5392
linux	1464
linux-source-2.6.20	1034
update-manager	826
linux-source-2.6.22	724
firefox	684
kdebase	673
firefox-3.0	668
ubiquity	590
openoffice.org	566

Bugs with the shortest titles:

select bug, title, length(title) as len
from bugs
order by len asc
limit 5;

bug	title	length
190560	-	1
160381	uh	2
224350	css	3
133621	gnus	4
138052	pbe5	4

If you want to play too, you can fetch the SQLite3 DB (5.8M, lzma-compressed), the DB creation script, and the script that fetches the bugs and import them into the DB. Comments about my code would be very appreciated (stuff like “oh, there’s a better way to do that in python!”), as I’m not very confident about my pythonic skills. :-) Update: apparently, I’m not really fetching all the bugs. I’m getting the same results as when you just press “Search” on https://launchpad.net/ubuntu/+bugs. But if you click on “Advanced search”, then select all the bug statuses, and click search, you get a lot more bugs (154066 vs 49031). If someone know which bugs are excluded with the default search, I’m interested! Update 2: Got it. Apparently the default search doesn’t list bugs that have all their “tasks” marked “Won’t fix”, “Fix Released”, or “Invalid”.

finally I managed to book my flight to DebConf8.

2008-08-03 IB3537
7:45 / Munich, Franz Josef Strauss, Terminal 1
10:25 / Madrid, Barajas, Terminal 4



2008-08-03 IB6845
12:25 / Madrid, Barajas, Terminal 4S
19:40 / Buenos Aires, Pistarini, Terminal A



2008-08-25 IB6844
21:45 / Buenos Aires, Pistarini, Terminal A
14:30 day +1 / Madrid, Barajas, Terminal 4S



2008-08-26 IB3564
16:30 / Madrid, Barajas, Terminal 4
18:55 / Munich, Franz Josef Strauss, Terminal 1

& I'm still surprised how frequently the prices for the same flights change ...

Italian weather agencies Most Italian regions have agencies providing good regional weather forecast. However, I could not find a useful national index of all their weather forecast pages. So I made one, tweaking an image map found on wikipedia. I only filled in those regions that I was checking to organise a specific trip. With time, if I'll keep using this map I'll add more regions. Of course, if you now of a weather forecast page of a regional, public funded agency that is missing here, send me an e-mail.

In May 2002 I had an idea for securing access to GNUPG [1]. What I did was to write SE Linux policy to only permit the gpg program to access the secret key (and other files in ~/.gnupg). This meant that the most trivial ways of stealing the secret key would be prevented. However an attacker could still use gpg to encrypt it’s secret key and write the data to some place that is accessible, for example the command “gpg -c --output /tmp/foo.gpg ~/.gnupg/secring.gpg“. So what we needed was for gpg to either refuse to encrypt such files, or to spawn a child process for accessing such files (which could be granted different access to the filesystem). I filed the Debian bug report 146345 [2] requesting this feature. In March upstream added this feature, the Debian package is currently not built with --enable-selinux-support so this feature isn’t enabled yet, but hopefully it will be soon. Incidentally the feature as currently implemented is not really SE Linux specific, it seems to me that there are many potential situations where it could be useful without SE Linux. For example if you were using one of the path-name based MAC systems (which I dislike - see what my friend Joshua Brindle wrote about them for an explanation [3]) then you could gain some benefits from this. A situation where there is even smaller potential for benefit is in the case of an automated system which runs gpg which could allow an attacker to pass bogus commands to it. When exploiting a shell script it might be easier to specify the wrong file to encrypt than to perform more sophisticated attacks. When the feature in question is enabled the command “gpg -c --output /tmp/foo.gpg ~/.gnupg/secring.gpg” will abort with the following error:
gpg: can’t open /root/.gnupg/secring.gpg’: Operation not permitted
gpg: symmetric encryption of /root/.gnupg/secring.gpg’ failed: file open error Of course the command “gpg --export-secret-keys” will abort with the following error:
gpg: exporting secret keys not allowed
gpg: WARNING: nothing exported Now we need to determine the correct way of exporting secret keys and modifying the GPG configuration. It might be best to allow exporting the secret keys when not running SE Linux (or other supported MAC systems), or when running in permissive mode (as in those situations merely copying the files will work). Although we could have an option in gpg.conf for this for the case where we want to prevent shell-script quoting hacks. For editing the gpg.conf file and exporting the secret keys we could have a program similar in concept to crontab(1) which has PAM support to determine when it should perform it’s actions. Also it seems to me that crontab(1) could do with PAM support (I’ve filed Debian bug report 484743 [4] requesting this). Finally one thing that should be noted is that the targeted policy for SE Linux does not restrict GPG (which runs in the unconfined_t domain). Thus most people who use SE Linux at the moment aren’t getting any benefits from such things. This will change eventually.

• [1] http://www.gnupg.org/


• [2] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=146345


• [3] http://securityblog.org/brindle/2006/04/19/security-anti-pattern-path-based-access-control/


• [4] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=484743

Share This

Managing a transition is not always easy, because many times they are bigger than you, and have parts you can’t directly take care of yourself because it’s not your area of expertise, you don’t have access, or whatever else. You end up, then, nagging people to do stuff for you, which for types like me is slightly uneasy. It is a pleasure, though, when you find supportive fellow developers on the other side, just willing to help you, even if you always have more work for them. Thomas and Michael, thank you! (And thanks, in general, to everybody who helps the release team, past and future. We just can’t do everything!) On reflection, I notice that I also find very gratifying to do stuff for others myself (particularly if I enjoy the task, of course), because it has an immediate payoff. Highly recommended.

Terah, Jacob, and I went camping last weekend. Terah had asked me what I wanted for Father's Day. I thought for just a second, and said "camping!" She looked like she regretted asking. But it has been a year since our last camping trip, so off we went.

We went to Marion County Park and Lake, a nice park built by FDR's CCC program in the 1930s. It is on the National Register of Historic Places thanks to all the stone buildings, bridges, etc. that were built back then and still survive.

The reason we like it is because one side of the lake has the park office, rentable meeting houses, RV hookups, a swimming area, and really all the things you expect at a park like this (albeit on a small scale). The other side of the lake has gravel roads and fire rings. No power, no running water, but the occasional stone outhouse every half mile or so. Which means that the RV owners are not going to want to spend the night on that side of the lake. So, unlike far too many parks, it is possible to pitch a tent a good distance from the RVs. It's so annoying to go to camping and find it to be noisier than home because of all the TVs and generators running.

Jacob wasn't so sure about the first evening, but he was having a blast all day Saturday. His favorite thing was the campfire. He called it "fie" (rhymes with "buy"). He watched me gathering up brush and preparing the fire, carefully placing the small sticks first, then the big ones. So he picked up a small stick, and carefully laid it inside the fire ring, too. Then he'd watch me light the fire, or blow on some embers, and he'd blow on it, too. Fffffffffffffff! As hard as he could. At one point, I was kneeling next to the fire ring, inspecting the kindling. Terah noticed Jacob kneeling next to the fire ring on the other side, with the exact same posture and expression as I had. It was a perfect Father's Day, I think.

Once the fire got going, he'd walk up towards it, say "HA!" (hot), and then run away. Educational, this camping business. He now knows what "hot" is all about.

Jacob also loved the lake. We would walk the few feet down to the beach, and he'd (holding on to one of us) walk just far enough into it to get his feet covered by the water, then pick up sand and throw it into the water to watch it splash. Or, he'd point at a boat ("bope") in the water ("wah") going past, all excited -- Bope! Bope!!! BOPE!!!

One of my favorite things while camping is sitting outside and watch the sun set. Terah and I did that while munching on banana boats we cooked in the fire (a banana and chocolate concoction that you sort of melt in tinfoil). Or we could sit in the shade on the ground and just relax and enjoy the scenery.

Terah didn't say that she enjoyed camping, but she did agree that there were good times and we could come back. Woohoo, I guess.

We're not really camping pros, and we learned a few things. One is that it would be nice to not have three meals in a day that require a fire. I'm fine with eating fruit, canned stuff, and other no-cooking-needed stuff when camping, but Terah likes more traditional food. And that's fine, and we had some great stuff to eat, but building fires is hard and hot work.

Over the first few days after we got home, Jacob came up to me and said, looking rather sad, "Fie? Fie?" I said, "Jacob, we don't have a fire here. There is no fire here." He would look so sad, and just repeat to himself for a few minutes, "No fie. No fie. No fie." So I asked him if I should build a little fire. "Yeah!" So I got a match and lit a candle. Oh, was that exciting, and almost as soon as it showed up, he tried to blow on it. By today he has learned how to blow out a candle.

A great Father's Day for sure, but we hadn't really tried to sow the seeds of a fascination with fire on the trip.

(Is this a network?) Terry Lane asked:

"Do you know of any reason why someone would suggest we consider placing blogs on more than one server? I think his main concern would be related to SEO and - I'm assuming here - Google."

There are some small reasons, mainly about various sorts of reliability: what if the server catches fire, what if someone filters out adverts from the server, what if the server gets labelled as a spammer or splogger, and so on... but I believe they're outweighed by ease of management and having all your site on one server. If the search engines label your IP address as a spammer, you can get another IP address as a short-term fix, but in general, the search engines are always a major threat to a blog-based business. If they label one server as a spammer, I think it would take the guys at google or whatever about 0.1 seconds to spot the link to another server. The best tactic is to avoid looking like a spam source in the first place... I guess if you're hosting several blogs on shared servers, like I do, splitting your blogs across several servers is a good idea for those reasons, which is part of why I do it... In any case, make sure you download backups in case your hosting goes like this:- The Planet (EV1) Data Center Catches Fire - 9000 Servers Offline If a blog becomes really popular, the usual tactics of mirroring and distributed load-balanced hosting can be used, but I don't think that was the question here. Even after all those, I can't think of any real killer reasons to split a blog across multiple servers if you're on your own dedicated server already. Have I missed a reason?