ZODB Talk Report
Christopher Lozinski <[email protected]> Tue, 31 Oct 2017 08:21:11 +0100
| Newsgroups | gmane.comp.web.zope.zodb |
|---|---|
| Message-ID | <[email protected]> |
My talk is getting better. After my first talk, they asked me who is using it. So now I talk about Plone and Brazil. After the second talk they asked about indexing. So I talked about repoze.catalog. Referring to the previous emails, certainly on indexing, or reindexing, the objects are activated and so one should not need to supply an object id. Now the questions are getting much more intense. The interest level is increasing. How does the ZODB clear the cache? "Probably when it needs more memory, it clears out the least recently used. All Python, so one can easily change the logic. I never have had to worry about it." They are asking about Abstract Base Classes, and Zope.interface. Which should they learn? I need to search for a good comparison, although clearly they do different things. ABC’s say you have to implement methods or attributes. One can check if they have been implemented. One cannot instantiate a subclass until they have been implemented. Whereas Zope.interface provides meta information about those methods and instance variables, supports CRUD and supports the whole view lookup machinery. I do talk about using other databases. I should also add a slide about other Python Databases. http://opensourceforu.com/2017/05/three-python-databases-pickledb-tinydb-zodb/ <http://opensourceforu.com/2017/05/three-python-databases-pickledb-tinydb-zodb/> And still no-one is starting to use it, because of my talks. One guy did come up to me later in the conference. He is very interested in doing roughly what I am doing. A directory of Python books. He wants the better books to have larger images using graphics libraries. His business model would be affiliate marketing, he would get some money every time someone bought a book on amazon.com <http://amazon.com/>. Although no mention of how to get traffic. He did not want to say much. He did say that when he searched for information on the ZODB, read the docs said that write performance was slow. I believe that the ZODB can now write to multiple files simultaneously, so if there are such references, they should be changed. He did not provide a URL. I should add a slide on this. I should add a slide on the testing story, although first I need to find some good docs on that topic. It is also interesting that for my first talk, I had way too many slides. It was not clear which ones to throw out. Now it is getting clearer. I just tossed out another 3 slides. I got rid of three of the slides about my application.The newest release is at: https://pythonlinks.info/presentations/zodbtalk.pdf <https://pythonlinks.info/presentations/zodbtalk.pdf> It is also getting clearer on what else I should add in. There are a bunch of Python conferences happening around the world. I invite others to give the talk there. At least it gets you a free ticket, which usually includes very nice lunches, and the contributors dinner. Warm Regards Christopher -- You received this message because you are subscribed to the Google Groups "zodb" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.