3. October 2012 19:56
I am now on my way home from the last day at GOTO and it has been another day with lots and lots of input. I chose to spend the afternoon watching two talks from the HTML5 track, followed by one on the humans at work and then lastly seeing the closing keynote.
The two talks on the HTML5 track were by Scott Hanselman who talked about mobile development and Dan North who declared “The browser is dead… long live the browser”.
The red line through Scotts talk was that there are different paths to take in order to provide a better experience for the rapidly growing number of users who are using a mobile device. With some countries actually being pretty much exclusively mobile users it is very important to think about this. As it is said, not making a decision is also making a decision, but you are really hurting a large percentage of your users if you choose to ignore the mobile world.
Dan Norths talk revolved around 4 key points. Everything is asynchronous, the DOM is the template, the server serves data and someone has already solved it. Building on these points Dan demonstrated a simple architecture using NodeJS to serve data and everything else done in the browser using some of the popular frontend frameworks like jQuery, Knockout and Twitter Bootstrap. Once again Dan really caught the audience, got some good laughs on the way, while getting his points across.
Liz Keogh’s talk on the humans at work track was entitled “To be honest”. Her talk consisted of a range of stories from real live, about how different kinds of lying gets us into trouble when building software and coming up with solutions. Estimates, scrum, the agile manifesto, bug counting and talking about something as being done are all lies. So in order to progress we need to get better at communicating and being honest. She pointed out that “you should not let your process hide your ignorance”. Processes tend to force us to do things in fixed formats, which in turn make up subtle lies and cloud honest and clear communication.
Finally it came time for the keynote, where Anders Hejlsberg went over which problems they aim to solve with TypeScript, while covering some of the design choices it has been based on. Naturally part of the talk was overlapping with what I had seen the day before, but I am glad that I stayed the the talk because he did have some deeper samples and just overall gave a great talk.
3. October 2012 13:10
I just finished my second and last interview of the conference, where I was lucky enough to get to speak to Anders Hejlsberg. It was a really good experience, and even for a guy who is just playing raporter for a few days it was really comfortable. I have no doube been nervous about doing interviews, not as much because of the rock star status that Anders and Scott have, but the whole situation is just completely new to me. In both cases it was just a matter of getting started, and then the calm that they posses just takes over.
Interviewing Anders today was really cozy and right off the bat just felt like a good conversation with a good friend. So I am hoping that my recorded material lives up to the experience, because right now I am thinking about when I can find a few hours for editing. And I am hoping that Trifork are happy with how the introduction of bloggers has worked out, so maybe I get the chance again next year.
As for the program today it has been a bit of a soft start. the keynote about the large hadron collider in Cern was mindblowing, but also so much out of our field that I don’t really feel like as got as much from it as I have with the other keynotes. After that I grabbed 3 lightning talks on making hard things easy. All 3 were great speakers, and especially Dan North managed to get some good laughs. At the same time it confirmed to me that lightning talks are difficult to do in a way where you really get your points across while having just enough debth. Yesterday I actually promised to do a lightning talk myself at our local usergroup, so I am hoping I can use some of what I learned from watching these really skills speakers get the most out of their 15 minutes.
Now – more talks, with both Scott Hanselman and Anders Hejlsberg taking the stage this afternoon.
2. October 2012 22:08
Today was the second day of GOTO, and it has been a really fun and interesting day for me. As I wrote yesterday I got to do an interview with Scott Hanselman. Actually it almost fell through, because of a communication blip, but thanks to Twitter the day was saved. Scott found out that I tweeted about doing the interview, so he tweeted back, we made the arrangements and got back on track.
As it turns out doing an interview with a guy who knows so much about doing interviews himself was a great way to start. Scott gave me a couple of pro-tips reguarding sound and editing, and he had a set of microphones in his bag that we used to make the quality even better. I won’t give away any spoilers today, but once we get past the conference days I will write about the actual interview and work on the video.
In reguard to talks I went to a talk on Riak Core with Steve Vinoski which was really good and he managed to convince me that I need to take a look at Riak as an alternative to where I have been leaning towards mongoDB. The distribution model in Riak seems very well thought out, and it seems like a good product to work with.
Later I gave in to my interest for programming languages, with a talk by Damian Conway entitled Sex and Violence about Pearl 6 and then a talk about polyglot programming. Finishing off the talks I saw Hadi Hariri from jetbrains talk about developers as the primadonnas of the 21st century. I think this is the kind of talk that we as developers need to see once in a while. Hadi made some good points about all the stuff we do that makes us lose focus. Methodologies, development tools and principles drag us into dogmatic wars on practices causing us to drift away from delivering solutions to real programs and away from programming motherfucker!
1. October 2012 23:35
Now the evening is comming to a close, on what has been a really interesting day. Since my last post I went to 4 very different talks on Raspberry PI, Myths of project management, No SQL and lastly Damien Conway on Contemporal Virtual Nanomachine Programming in Topoligically Connected Quantum-Relative Parallel Spacetimes – made easy! So I got to get around very different aspects of life in the software industry.
Conways talk was no doubt the most crazy talk I have ever seen, and I don’t even want to know how much time has gone into preparing that talk. I can safely say it is the geekiest moment of the conference so far, and I am sure my brother in law who works with nuclear physics would have laughed his ass off as much as I did.
After the talks were done I went around and visited some of the stands, where especially the MongoDB stand cought my interest. I have really enjoyed working with MongoDB and I got to hear more about what they are working on, so that was brilliant. A key point for me is that they are doing full-text search in vNext, making it a lot more interesting in the search space that I recently released a whitepaper on.
The evening was then spent at the conference party, where I sat between two speakers, Steve Vinoski and Matt Heitzenroder. Matt was in the NoSQL panel that I was at earlier, and Steve is a legend for the guys in the Erlang and Riak space. So I had a really nice evening hearing about how they experience a conference, and the challanges they have.
Tomorrow I will probably be blogging a bit less I think, because I will be preparing and doing my first ever interview. I could not ask for a better opportunity as I will be talking to Scott Hanselman, which means I will start out interviewing the speaker and blogger I respect most of all.
1. October 2012 13:24
After Kaspers talk I stayed for a talk about dynamic languages in production. It was a very energetic performance, and a fun talk for a pretty lowlevel subject. The central topic of the talk was debugging dynamic languages in a production environment. This has historically been very hard, but they have made some strides toward making it possible in Node, through the analysis of dump files and looking instrumentation of applications.
So for now it has been really interesting, but I will be hoping for more takeaways that I can use in my everyday job. I thought there would actually be more of that in this track, but it has not been the case so far..
1. October 2012 10:56
As with most conferences GOTO started with a program talk followed by the first keynote. Today’s keynote was by Rick Falkvinge from the pirate party about “The red flags on the internet”. It was a very interesting keynote comparing some industries of today with industries that have died out because they refused to keep up with technological advances and tried to protect their own interests by limiting the interests of the public. The comparisons went back to the catholic church and governments who tried to limit the printing press, to the railroad and stage coach who were behind the red flags act of 1865.
The red flags act said that any horseless carriage should have at least a 3 person crew, on of them walking in front of the automobile waveing a red flag. More recently a comparison was made to Polaroid and Kodak who revolutionized digital photography, but ultimately went bankrupt, arguing that they died out because they ignored technological advances and took a protectionist approach.
Falkvinge went on to make some predictions about which industries will suffer the same fate. The obvious industries being postal services, news services and telco, who are under a lot of pressure due to technological advances, making them effectively obsolute. In the more far out predictions department he predicted the demise of banking due to crypto currencies and even governments do to technological advances that allow government to be done by the public.
The overall point of the talk was that we should protect the internet as it is today, and that it is in no way a new tendency that companies and governments try to limit the publics access to information and new technology. Throwing tantrums in public to get tax payer money for protecting against some theoretical danger we are supposedly facing is a pattern we see all the time. So it is probably something we will continue to see, but it is worth being aware of.
It was fun to see a talk that I think would be pretty controversial in some places, but probably not as much at a developer conference.
28. August 2012 17:37
Sorry but I just can’t help myself when a crazy sentence like the headline presents itself and actually makes sense. None the less it is a question I have been asking myself each year while I have been living and working in Aarhus. Every year something has kept me from going to GOTO – mostly some combination of the attendence fee and bad timing of other things I have been doing. This year I had kind of dismissed it having already gone to TechEd, so I figured my budget for conferences was spent.
Luckily I was offered to go to GOTO for free as a blogger by the kind folks at Trifork. More...
22. June 2012 20:47
I had the pleasure of going to TechEd North America 2012 in Orlando. With a lot of interesting news comming out of Microsoft it was 5 very exciting days. While I was away I did some blogging on the company blog (in danish).
As you might expect Azure and Windows Metro style design were the big topics of the conference. With Azure finally taking a direction that I think makes it interesting, and Windows 8 in general looking like the next milestone for Microsoft it was really good.
26. April 2012 14:30
Thanks to everyone who participated in the first ANUG lightning talk this wednesday. It was a fun to dive into the subtle differences between Nancy, ServiceStack and WebAPI. As with everything else – it is always more fun when you win the battle of course :)
The demo and slides are available at dotnetnerd.dk/talks/webapi.zip - along with a small sample of using type providers in F# for building a WebAPI service.
4. April 2012 10:35
Later this month on the 25th I will be speaking along with two other speakers about webframeworks at an ANUG meeting. The frameworks covered are ServiceStack, Nancy and WebAPI where I will cover the latter.
While I have been preparing my eye cought a new feature in F#, which I think is one of the first really good stories for F# that makes sence outside akademia and science. Type Providers give us a way to access data sources that provide its own metadata, and this fits nicely with building WebAPI services. More...