tag:blogger.com,1999:blog-7831813422886730737.post1198246112237623163..comments2023-10-08T10:44:28.524+03:00Comments on Event Processing Thinking: More on the two types of sliding windowsOpher Etzionhttp://www.blogger.com/profile/10791357917675270335noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-7831813422886730737.post-22835805319498227662011-06-15T13:51:52.011+03:002011-06-15T13:51:52.011+03:00Hi Marco.
I'll post a follow-up posting to ...Hi Marco. <br /><br />I'll post a follow-up posting to answer your question, the main key is the events may have various distinct roles. <br /><br />cheers,<br /><br />OpherOpher Etzionhttps://www.blogger.com/profile/17070103285719046013noreply@blogger.comtag:blogger.com,1999:blog-7831813422886730737.post-22603516337276228532011-06-15T11:35:43.765+03:002011-06-15T11:35:43.765+03:00Dear Opher,
thanks for replying to my comment on ...Dear Opher,<br /><br />thanks for replying to my comment on your previous post. I am, however, not completely convinced that the first point you are making is valid.<br /><br />If there was an event at 10:35, the window would already close at the occurrence of this event. Using the same course of argument, I could say that an event occurring at 10:15 would be included in your closed interval semantics, therefore the window would contain more than 5 events.<br /><br />Considering the point in time, when events are generated, you are completely right. This is different between both window semantics and this difference might be a major one, depending on the application type.<br /><br />So, thanks for your clarifications!<br /><br />Warm Regards<br />MarcoMarcohttps://www.blogger.com/profile/08804161575647221766noreply@blogger.com