Profile cover photo
Profile photo
Markus Völter
1,328 followers -
software, podcasting, gliding
software, podcasting, gliding

1,328 followers
About
Posts

Post has attachment
Ok, so here is the final set of photos from Gap this year. I took 450 pictures, most of them crap, but those here are reasonably ok. Have fun. And go get a glider license, so you can do that yourself. It's worth it :-)
PhotoPhotoPhotoPhotoPhoto
2016-09-11
32 Photos - View album

Post has attachment
Photo

Post has attachment
Last Wednesday I flew 937km, the farthest I ever flew. It's interesting to dissect why I flew that far, and whether I could have reached 1000km, the "magic" number for glider pilots.

http://www.onlinecontest.org/olc-2.0/gliding/flightinfo.html?dsId=4974916

The first question is trivial to answer: the weather was unbelievably good. We had fresh cold air, which means we had strong thermals (I had several over 5m/s). However, unlike, typically for cold air, we also had a high cloud base (which means you have a lot of altitude to work with): it was over 2000m most of the time, sometimes up to 2800m. Also, and maybe most importantly, the weather was very homogeneous: 2-4 eighths of clouds all over southern Germany. Even those areas that are often blue or otherwise "slow" (such as the Kraichgau) had very nice clouds. For an objective measure for the quality of the weather take a look at the OLC page:

http://www.onlinecontest.org/olc-2.0/gliding/daily.html?st=olc&rt=olc&df=2016-05-04&c=DE&d-2348235-s=4&d-2348235-o=1

Last I checked, it showed 36 flights of 1000km or more. I think this has never ever happened before. So the weather was really exceptionally good.

The fact that there were 36 flights over 1000km or more also answers the second question: objectively, it was possible to fly the magic number, and the real cracks did it. So why didn't I?

Let's first look at what went right in my flight. I flew an average of 117,4 km/h track speed over the 937km. Considering the long distance, this is really fast. In fact, it is the fastest average speed of the whole day in all of Germany (ignoring 4 faster flights that were only half as far -- much easier to fly fast over a shorter distance):

http://www.onlinecontest.org/olc-2.0/gliding/daily.html?st=olc&rt=olc&df=2016-05-04&c=DE&d-2348235-s=6&d-2348235-o=1#p:0;

So the flying itself was good. I also noted that I didn't have one of my typical "down times": when I fly for 8 hours or more, I typically lose concentration at some point and start making mistakes, such as flying with the wrong flap setting or failing to find or center a thermal.

So why didn't it work out? There are several reasons. First, we should have taken off half an hour earlier. The group of pilots at Messelberg were spoiled by the locally very low cloudbase in the morning. A few km south it was already good. We should have seen that and launched earlier. Second, I maybe should have taken water ballast (which makes the airplane faster). I decided against it because I was going to fly in below zero temperatures, which means that the water may freeze into the wings. I had that last year, not good. Third I should have been a bit more patient in the evening, trying to fly a bit farther. Because of the intricacies of how the scoring works, I would have had to fly about 100km more to "get" the missing 63km. Looking at the flight of my buddies from Giengen ...

http://www.onlinecontest.org/olc-2.0/gliding/flightinfo.html?dsId=4969746

... it turns out that I could have tried to join them on their late-day trek to Dinkelsbuehl and back. That would have done it. Finally, I should have turned in the north at Wasserkuppe and not at Suhl, that would have avoided the sub-optimal "kink" in the leg to Beilngries. A final (meta-) mistake was that, in the morning, I didn't even think about trying for the 1000km. If I had, maybe I wouldn't have made some of the other mistakes.

So, are 1000km possible (for me) in the future? If the weather is perfect like on that day (which doesn't happen often!), and if I really try, I think I can make it. This realization is the relevant learning of 937km.
PhotoPhotoPhotoPhotoPhoto
2016-05-06
10 Photos - View album

Post has shared content
The MPS Language Workbench Vol I, third edition, updated for MPS 3.3, is now in Google Play https://play.google.com/store/books/details?id=nvcEAwAAQBAJ
Previous editions have been updated, so make sure that you sync your library to get the new content.

The free book preview, recommended reading to get started with MPS, has also been updated. Available from http://books.campagnelab.org
Photo

Post has shared content
Get the MPS user guide as a nice looking pdf thanks to Marco Lombardo. http://goo.gl/kraCMD

Post has attachment
Introducing Grammar Cells for MPS

MPS projectional editor affords users a lot of flexibility in terms of notational flexibility and language composition. I have posted many examples of this over time. 

However, this comes at a cost: building editor in MPS that don't just look nice but that also are nicely usable can be challenging. In particular, editor developers have to create so-called actions -- wrappers, side transforms, substitutions, delete actions -- that, for example, allow entering expression tree structures linearly. Enabling the insertion of parentheses in cross-tree locations, or splitting a number literal in two is even more challenging. Implementing all of this correctly can be a challenge, even for practiced users of MPS. More importantly, it is easy to forget some of those actions, which then leads to editors that behave inconsistently: good in some cases, and bad in others. Such inconsistencies are a major source of frustration for new MPS users.

To address this issue, we (mostly Sascha) have built the grammar cells infrastructure. These are new editor cells that have enough semantic richness so that the necessary actions can be generated automatically. To implement a nicely usable expression language, developers do not have to write a single line of action code. Check out this video:

https://www.youtube.com/watch?v=eANAxKMNBAU

Even though language developers have to invest a little bit of effort into learning how to use these new cells, it is certainly simpler to work with than the actions. In particular, it is easier to get things consistent. Check them out, they are in the master branch.
Photo

Post has shared content

Post has shared content

Post has attachment
Layers.
Photo

We are still looking for new people in the mbeddr(MPS team at itemis. If you want to do interesting work with cool technologies in a fun team, ping us: http://mbeddr.com/2015/07/28/weAreHiring.html
Wait while more posts are being loaded