[translated] Thank you for your question. One thing that I would like to make clear is that for XNGP when it’s used within a city, an urban scenario, it’s actually quite different than Highway NGP when it’s used on a highway setting, especially within the cities we are talking about, areas without high definition maps. It’s very hard to really judge how much ahead in the competition we are compared to our peers because right now within the industry, we only see a couple of companies that are able to claim that they have similar technologies, but they are really limited to scenarios where there is high definition maps available, whereas for XPeng, we are actually able to roll out our XNGP software in areas without high definition maps. We also have the capability to actually roll out it further to 50 to 200 cities, so that is definitely very, very advanced. That’s the only thing I can say. Typically it takes about 12 to 18 months to be able to use the XNGP [indiscernible] functionality from the testing phase of the greyscale testing to the actual rollout, so it’s quite challenging to reach that milestone of the actual rollout. I believe that definitely for XPeng, we have very solid technology in our architecture that actually supports multiple sales cycles, including data cycles within the XNPG software itself, and so I would say XNGP on average is leading ahead the competition by about 12 months time. The second part of your question is about the significance of data to the actual performance of the technology. Now, I would say data definitely means a lot and it’s very valuable to autonomous driving; however right now, based on our current sales scale and adoption rate and also the deliveries in the near term, we believe that data here is not going to cause a big issue or really pose a lot of difference, because right now we are still doing vision-based or vision-centric data gathering, and in the future we expect to have actually more data coming from, for example, language positing and language input, and that will also help us to actually have better performance in the software execution. I think that right now, our capabilities to support that kind of data processing is proficient. Definitely in the longer term, data will really mean more when you have more autonomous driving cars on the road that can actually contribute to your data set, but it’s in a much longer future, and so right now it’s very hard to say where or when the inflection point will be. Thank you.