I have already visited the US in June, when I attended the Plone Symposium in New Orleans. On that occasion, I stayed in the US for ~30 days. This time around I was planning to bring my wife and dog (the cute Yorkie pictured at the top of this blog).

One thing that worries me though is that I’ve never took my pet on any flights, so I have no idea how pet-friendly airlines are. Just to get from Sao Paulo to the US (Houston) it’s a ~9:30h flight. My dog is usually cool with traveling. Earlier this year I drove nearly 8h to the beach and we stopped twice on the way, but she did not pee until we arrived at the destination.

So I’m putting this out there, looking for any feedback from people that took their pets on flights.

As for myself, I really loved meeting some long-time friends that I hadn’t seen since PyConBrasil 1, two years ago. After the event I had the opportunity to hang out with Nate Aune, of Plone4Artists fame. We drove from Joinville to Florianopolis, stopping at the Parque Unipraias in Camboriu. View Map

And then, the next day we did some sandboarding at the Joaquina Beach in Florianopolis. We had such a great time. Finally around 9pm we’ve dropped Nate at the Hercilio Luz Airport in Florianopolis and finally headed home, arriving at 4am, completely exausted. View Map

My talk was about my Google Summer of Code project, “Improved out-of-the-box WebDAV experience”. But ended up being more about the Summer of Code than about my project.

Slides available below. It’s a mix of Portuguese and English, so you might actually be able to make sense of it. :)

On my next post I will talk a little bit more about PyConBrasil itself…

Sunday, night train to Vienna. Lecture on WebDAV at PloneConf monday after
lunch. ZopeMag is kindly publishing the article for free (layout
still unfinished at this time). Hanging out with the Zope
demigods. Meeting Martijn Faassen, Jim Fulton and a couple others for
the first time.

Thursday, leaving to Brno, staying at the house of some guys from
#zope-cz. Friday in Praha. Saturday in Dresden, visiting philiKON
(gotta get started with reviewing that book sigh). And then visiting
Berlin on Sunday-Monday, possibly staying at philiKON’s parents.

Daiane, my wife, is coming with me. And so is Fabiano, my ex-coworker
and long-term friend. He’s never been to Europe before, and he’s also
giving two talks on PloneMall, at the PloneConf and OSCOM.

In Stuttgart, we decided to walk around a bit during the nite to get
used to the city, and to wait for the next morning to visit
BodyWorlds. We had another Donner, and then Alan decided to go to an
internet cafe to check the address and times for BodyWorlds. To his
(and my) surprise, he was wrong about the city. BodyWorlds was in
Frankfurt, and not in Stuttgart as he thought. So, we decided
to do it the right way this time, and plan every step beforehand. I
had my flight departing from Zurich on the Tuesday, 10pm and Alan
wanted to go Amsterdam to hangout with Lon, from Q42. We checked
all the schedules to make sure we would be able to get to our
destination at the right times, and on the next morning Alan bought
the tickets for us to Frankfurt, plus my ticket from Frankfurt to
Zurich, and his from Frankfurt to Amsterdam, as well as make the
reservation of his flight from Amsterdan back to Houston.

Time proved that planning is always the best thing to do, and
everything went as planned. We got to the BodyWorlds with enough time,
and we even had time to visit Andy Warhol’s Time Capsules on the
Museum fur Moderne Kunst in Frankfurt. Actually we didn’t even knew
that it was open during these days, but accidentally found ourselves
in front of the museum during a walk (while trying to find nice gifts
for the girls at home).

From the museum, we went straight to the Haupftbanhof, where we had
left our luggage on the lockers, about 15 minutes before my
departure. Alan’s train would leave one hour later, so, time to say
goodbye.

After all, I think it was an amazing trip. If wasn’t for the fact that
we didn’t plan well after leaving the sprint location, it would had
been perfect. I’m looking forward to the next one!

I would like again to thank Jodok for his excellent planning of the
sprint, and all the sponsors who made money available for the sprint to
actually happen. I would also like to thank Alan for having paid a
good amount of my expenses after we left the sprint (I didn’t had
enough money with me, unfortunately).

Oh, and I almost forgot. There are a lot of photos I took during
the days I’ve been there, thanks to my mother, which allowed me to
borrow her brand-new digital camera.

Interesting things I’ve learnt during this trip:

• German uses a stylized ‘B’ (which reminds me of the greek ‘beta’)
  which on spoken word sounds like a double ‘s’ (ss). They even seem
  to write it as a double ‘s’ when the letter is not available (maybe
  misconfigured keyboards?). So, "Domstraße" -> "Domstrasse".
• Fanta outside Brazil is yellowish for some reason. I asked
  around and it seems it’s like that everywere, except in Brazil,
  where it’s more like bright orangish.
• Zwellings -> Twins.
• Lots of other german words that I forgot in the course of the last
  two weeks :(
• Donner Kebab is a really nice food, and it seems to be from turkish
  origin. It’s made of lamb meat. The more pepper, the better.
• Hum. Just found out about Googlism. Check out What is Plone,
  and What is Zope

What happened next is that we took the train to Praha, only to find
less than 20 minutes later the hardest way that brazilians need visa
to enter the Czech Republic. What a mess. We had to jump off the train
on the next station and take the train back to Vienna. It was really
cold outside, and getting dark. From the time we left Vienna to the
time we got back it was nearly four hours. Ugh. Alan seemed quite
pissed at me. Alan called Robert Boulanger which promptly found us a
nice hotel next to the Westbanhof. The hotel was amazing, with bathtub
and all, and it was really nice taking a shower.

Next morning, off we go to try to get a visa, but again luck wasn’t on
our side. It was sunday, and the Czech embassy was closed. Alan
quickly decided that we should turn around and go to Germany
instead. He wanted to see the BodyWorlds expo, which he was
sure was in Stuttgart. There we go, another train. We had to
change trains in a little town (which I can’t seem to remember the name, I
guess was Leinz or something), and the next train would be around two
hours later. We walked around, trying to find something to eat, but
there was only a McDonalds open. Ugh. A few more steps and I devised a
Donner place (suggestively called Food Master). It was quite some
luck, because it was Sunday, and every single shop was closed, except
for the Food Master and McDonalds. During the food, I and Alan
discussed the peculiarities of the government and the expectations for
the future of both US and Brazil. What a chat!

Time for calling home, notifying the girls of our messy situation, and
to take another train. We arrived in Stuttgart when it was already
dark, so the first thing we did was to find a hotel to stay. There
were a couple ones not too far from the Haupftbanhof, so choose one of
the first we found.

On the train from Leinz (?) to Stuttgart, Alan and
I were discussing ways to make Plone faster, so we profiled a bit and
figured that quite some time was spent on rendering templates,
specially loops and stuff like that. Some stuff was being calculated
again and again for each item in the folder_contents template, for
example, and that was quite bad. So, I decided to use a helper View
instance that would be used for doing the expensive calculations, and
would be available on the context of Page Templates. It was quite
quick to hack it, and after some profiling we had it running in with
an improvement of 8s over the unoptimized template (22s original, 14s
optimized). Then I fiddled with it a bit more to make the View use the
Ram Cache machinery, and we cut on another 3s. At this time, the
optimized template was running roughly 10s faster than the
original. Alan seemed pretty happy, given our bad luck on the trip
itinerary :)

(to be continued…)

I’ll try to make it short. Anyway, I don’t remember the exact order in
which things happened during that week. So, basically, I and Godefroid
went down the mountain once on the snowcycle, and on the second
round we found Jodok and the guys from Burning Man. Godefroid took the
easy track, while we took the ski track. Everything was reasonably
under control until the middle station, when we decided to continue on
the ski track instead of taking the elevator. Then things turned very
bad. At some point I thought about giving up on trying to ride the
snowcycle and waiting for someone to rescue me. It was getting dark. I
couldn’t see the bottom of the mountain. Godefroid convinced me to
continue and we kept going until the last few hundred meters, when I
just gave up and walked.

I think I’ve never been so tired in my whole life. My whole body
was hurting. My hands were shaking. I had to move very slowly for the
following 3 days. Not bad for the first time in snow I must say. :)

This same day Andy arrived at the sprint location, and it was
really nice meeting him, specially because I had worked with him in
the past, but we’ve never meet personally. A funny incident
happened. Andy confused me with Kapil, until I took my hat off. ;)

During the next days we made lots of progress in all fronts. There
were some short seminars during the afternoons on several topics
including i18n. I gave a short lecture on Archetypes, and demoed the
Zope3-based application I’ve been developing together with Alan.

There was some discussion about a installer tool with Andy, Kapil,
Philip (Auesperg) and me. We decided that it would be nice to have a
XML-based format for representing package dependencies and
metadata. Kapil sent me a sample file, and I need to make ArchPackage
export something like that.

I was very interested in trying CMFMetadata, so after a short
introduction by Kapil, I did some polishing to make it usable. Then,
Kapil was asking me about ways of integrating CMFMetadata and
Archetypes. I was a bit unsure at that moment, but after some sleep, I
woke with the whole picture in my mind. It wasn’t very hard to
implement it. The solution goes more or less like this:

First, I changed the Schema class of Archetypes to have a better
handling of fields, moved some methods around, used a different
structure to keep field ordering constant. Then I’ve built a
CompositeSchema class, that takes a list of Schema objects, and
joins their fields together. Basically, it acts like a normal
Schema. Then, I’ve created a FacadeMetadataSchema class, that
takes a the Id of a CMFMetadata Set on the constructor, and builds the
Archetypes fields on-the-fly from that. I also created a
FacadeMetadataStorage that is used for these on-the-fly-created
fields, and that uses the metadata tool from CMFMetadata to retrieve
and store values (which on its side, uses the annotation tool from the
Annotations product, built by Kapil for Infrae).

So, by the time I finished this, it was almost 5am on the Saturday,
and everyone was going to leave on that day. Alan had invited me to go
with him to Praha, but I hadn’t decided yet. Then, Phil, the Prince,
suggested that we go with him to Vienna, and from there go to Praha,
and I thought it would be a nice idea, because I really wanted to see
Vienna again. In a bit more than one hour we
3 were on a train to Vienna. From that minute on, I was already
regretting of my decision to go, but little could be done.

(to be continued)

The organization was amazing. Everything was perfect, and
well-planned. We had a board just on the entrance with information in
general, like where would be the dinner, what time breakfast would be
served, what we would have for dinner, what activities were planned
for the day, and so on.

Around 50 people were present at the sprint location. Not everyone
arrived at the same time. I was the first person to get there it
seems, as I arrived on the Friday 6th at 6:40am. The first thing I did
was to get my luggage in a locker and go get some food. The bus would
only arrive at 12:30, so that would mean almost 6h waiting. Got some
food, bought a magazine, called Jodok to figure out who else would
arrive and what time and flight they would be on. Turns out that four
people were arriving on that morning: me, Sasha Vinscic, Leonard
Norgaard and Matt Hammilton. I had Sasha’s mobile number, so sent him
a SMS with instructions of where to find me. Then I’ve got a seat at
the airport shopping and waited for him, reading a magazine. Around
10:30, he found me, and we chatted a little bit, then called Leonard,
which had already arrived. By that point it was just about time for
catching the bus, so we headed to the bus stop. Only one problem, we
couldn’t find Matt, and no-one had idea of how he looked like. So we
just waited until he showed up.

The bus left Zurich, heading to the sprint location. The thing that
surprised me most was the clouds, which were completely disturbed
by the flow of airplanes. It wasn’t as cold as I would expect there,
but still cold.

So, after around 1h from leaving the airport, we crossed the border
between Switzerland and Austria, heading to Bludenz. I was amazed to
see the snow, which had completely covered the mountains. More amazing
than that was the fact that the weather was completely clear, with
some occasional clouds here and there.

Another hour, and we’ve got to Bludenz, where we then took a taxi to
Schruns-Tschagguns. Another 20 minute ride, and I was at the
‘Gasteshaus Davenna’, which would be my temporary home for the
week. The house was very comfortable and warm. Perfect time for a
shower. :) The only problem I’ve found was that the house maid didn’t
knew a word in english, so I didn’t understood a single thing that she
said. On the next day I discovered that she wanted to tell me that I
was in the wrong room, and had to change. Duh.

I called Jodok again to ask him for directions to the ‘Haus des
Gastes’, which would be the sprint location. He promptly said that
would pick me up at the door, and 5 minutes later he was there. The
Haus des Gastes is very close to the place I was, just a quick walk
away.

Getting there, we tried to setup networking, and surprisingly, the
other guys didn’t have any problem setting up their wireless
networking, except for me and Leonard. For some reason, the networking
wouldn’t work on Linux, but if I rebooted in MacOS X, it would work
flawlessly. Weird. A trip to the #debian channel revealed that the
problem was the ECN (explicit congestion notification) feature of the
kernel, and disabling it made the network start working
again. Unfortunately, Leonard didn’t had the same luck, and for some
reason this didn’t work for him.

Next morning the folks started arriving, slowly, and we started
working out issues. I was working on Archetypes, applying patches and
fixing bugs, to make a 1.2.4 release.

On the third day most people had arrived, and on the monday, we got to
the top of the mountain for dinner and then for some snow
sports. Pretty much everyone went for the tobogan, and some people
went for the snowcycle, which seemed pretty cool. I’ve gone twice for
the tobogan, and the second time was obviously better, as I discovered
some tricks to make it go faster.

The next day, Godefroid Chappelle invited me to be his partner on some
ski classes. I woke up pretty late, so didn’t had much time to
think. I accepted the proposal, which was made even better by the fact
that he offered to pay part of my class as a wedding gift. There we
go, up the mountain again. We got the ski gear, which seemed very
uncomfortable to me, specially the boots. The fact that they force you
to keep your knees half bent was a pain for me. In fact, I would
consider skiing my second preferred sport if wasn’t for that.

We were not so bad into learning, and, in fact, we could do the basic
exercises without much trouble. (And hit the ground only a couple
times).

If that wasn’t enough, we decided to go down the mountain again, this
time on the snowcycle. (to be continued…)

So, we decided to go ahead with the plan of converting the OO*s to II*s
and added a new feature to the plan, after a discussion over chinese
food: we would try to delay loading the objects until it was strictly
necessary. That would be possible because the objects are normally
fetched from a search result, and using only OIDs on the indexes would
allow us to return the object, given a OID when the user iterates
through the search results.

So, the workflow is more or less like this now:

• User does a query
• Catalog delegates query to the indexes
• Indexes returns a list of OIDs (actually a IISet)
• Catalog builds a Result object with the intersection of the
  OIDs received from the Indexes
• Result, when asked for an item, does a lookup by the OID and
  returns the actual object.

Needless to say, the improvement was overwhelming. Not only the query
was blazingly faster, but the database, after replacing the indexes,
was 20% smaller.

I must admit: the BTrees package its one of the most amazing ones I’ve
used during all the time I’ve been involved with python, and when you
deploy it the right way, it can make a world of difference.

This is not a general purpose ZODB sync, but instead, its very very
specific to the IndexedCatalog. So, we can actually call it a ‘Shelf’
synchronizer.

A Shelf is the ‘unit’ of work in IC, which contains multiple
‘Catalogs’, and each Catalog has a set of ‘Indexes’, one for each
attribute you want indexed. Indexes are basically a BTree, where the
key is the value of the indexed attribute, and the value is the object
OID (after our last changes, that is. before that the value was a
reference to the object).

It was very nice to work with them, specially Kiko. We had long
discussions about how the said synchronizer should be implemented, and
I think that the docs reflect our goals very well, now its wait and
see if it will actually work in practice (which I’m particularly sure
will be true), and looking forward to use IndexedCatalog myself!

I just regret cause I didn’t had time enough to look at Kiwi, which
was on my TODO list, but I expect to be back there soon for a crash
course :)

They’ve got pretty far, by using XSLT (or is that XLST? I never remember) to generate a XUL UI from the
XML. It was pretty cool to see it.

So, I’ve got all my stuff checked in, and its pretty late here (about
4am), while in Brazil its a bit before midnight. I haven’t really got
used to the timezone change yet ;)

I’ve been also talking with Kapil about his subversive ideas on a
Subversion browser based on Plone. He showed me some nice stuff
already working, and I was really impressed with it. I mean, I didn’t
know such thing was possible to do, and its really really nice stuff.