Data Streaming

It is really useful program to learn about streaming.

It is really useful program to learn about streaming.

It is a very comprehensive course

It is a very comprehensive course

I think the content is very nice, but the projects has unclear instructions and the workspace can definitely be improved because it really impairs the learning experience

I think the content is very nice, but the projects has unclear instructions and the workspace can definitely be improved because it really impairs the learning experience

The course is great. Got to learn a lot with hands-on project experience. I'm giving 4 stars because mentor support could have been better. 1-1 chat availability and live troubleshooting with the mentors would have been great. FAQs for common problems would have saved time. Overall, I would recommend this program.

The course is great. Got to learn a lot with hands-on project experience. I'm giving 4 stars because mentor support could have been better. 1-1 chat availability and live troubleshooting with the mentors would have been great. FAQs for common problems would have saved time. Overall, I would recommend this program.

The first part 'Data Ingestion with Kafka' was a good introduction to the topic of data streaming in general as well as the usage of Kafka and technologies around Kafka for data streaming. The only downside of the first part was, that the used software versions are about 3 years old and the Faust library which was used, is not maintained anymore (i.e. a dead project). It still shows an example of how stream processing can be done, but at least the Faust part seems not of practical relevance anymore. Also some current features of Avro (e.g. enums in schema) could not be used because of the too old Python version used for the examples and the project (3.7). The second part of the Nanodegree (Streaming API Development and Documentation) was the worst course I have seen from Udacity so far. It provides some general introduction on how to use Apache Spark, but the little actual information this course provides is spread out over too many lessons, and the exercises are full of errors, which waste the time of people taking the course. If you follow the instructions and things don't work the way the instructions say, you need to figure out what the problem is and what you did wrong - but mostly the answer is: You did nothing wrong, but instead the instructions were wrong. Apart from that, the exercises are always explained in several videos, some of them quite long and with additional information, some shorter and reduced to the most important information. So all videos show basically the same content, but in some videos some additional topics are mentioned. Again this is wasting people's time. The additional information of the longer videos could instead just be separated out and showed in a separate video or text. Nobody wants to watch the same content several times just to figure out some additional details, which would be lost if you watch only one of the videos. In general I'm very disapointed by the quality of this Nanodegree. Considering that this is a paid (and not exactly very cheap) program, I would expect it to be up-to-date, make efficient use of my time and not be full of errors. After this experience I will really think twice before taking another program from Udacity.

The first part 'Data Ingestion with Kafka' was a good introduction to the topic of data streaming in general as well as the usage of Kafka and technologies around Kafka for data streaming. The only downside of the first part was, that the used software versions are about 3 years old and the Faust library which was used, is not maintained anymore (i.e. a dead project). It still shows an example of how stream processing can be done, but at least the Faust part seems not of practical relevance anymore. Also some current features of Avro (e.g. enums in schema) could not be used because of the too old Python version used for the examples and the project (3.7). The second part of the Nanodegree (Streaming API Development and Documentation) was the worst course I have seen from Udacity so far. It provides some general introduction on how to use Apache Spark, but the little actual information this course provides is spread out over too many lessons, and the exercises are full of errors, which waste the time of people taking the course. If you follow the instructions and things don't work the way the instructions say, you need to figure out what the problem is and what you did wrong - but mostly the answer is: You did nothing wrong, but instead the instructions were wrong. Apart from that, the exercises are always explained in several videos, some of them quite long and with additional information, some shorter and reduced to the most important information. So all videos show basically the same content, but in some videos some additional topics are mentioned. Again this is wasting people's time. The additional information of the longer videos could instead just be separated out and showed in a separate video or text. Nobody wants to watch the same content several times just to figure out some additional details, which would be lost if you watch only one of the videos. In general I'm very disapointed by the quality of this Nanodegree. Considering that this is a paid (and not exactly very cheap) program, I would expect it to be up-to-date, make efficient use of my time and not be full of errors. After this experience I will really think twice before taking another program from Udacity.