Erik on Software

Sign in Subscribe

Topic

community

A collection of 47 issues

Community Comment: Part 33 - With respect to Databricks vs Snowflake, actual usage is more interesting than account creation

* Snowflake vs Databricks account penetration * Account growth rates not divulged * Data can be easily misinterpreted * Databricks: AI advantage, closing in elsewhere

Community Comment: Part 32 - Citizen users of SQL need multiple skills

* Citizen users of SQL need multiple skills * Understand what tables & views needed * Generate correct output * Performantly generate correct output

Community Comment: Part 31 - Wholesale self-service analytics is a dumb goal

* Dumb goal: wholesale self-service analytics * Don't attempt a shotgun approach * Self-service users need to be defined * Identify & cater to power users

Community Post: Part 4 - A book on deepfake detection quoted me

A well-received community post I recently made: https://www.linkedin.com/posts/erikgfesser_spark-bigdata-analytics-activity-7130307566965776384-ZYnv I'm apparently being quoted in books now. The quote is in chapter 3 ("Deepfakes Spark Implementation for BigData Analytics") of a book entitled "Handbook of Research on Advanced Practical Approaches to

Community Post: Part 3 - Delta Lake 3.0 is like an electric car charging standard

* The power of open source * Delta Lake 3.0 is just one example * Parquet-based data lakehouse file format * Similar to electric car charging standard

Community Comment: Part 30 - With respect to Azure, Databricks Unity Catalog & Microsoft Purview is not an either-or proposition

* Databricks Unity Catalog & Microsoft Purview * Not an either-or proposition for Azure * Likely need to integrate these products * Integration dependant on several factors

Community Comment: Part 29 - Usage of normalized versus denormalized data storage is dependent on usage patterns

* Normalized versus denormalized data * Data storage dependent on usage patterns * Denormalized data makes reads efficient * Normalized data makes writes efficient

Community Comment: Part 28 - Technical debt is actually a subset of organizational debt

* Not all technical debt is the same * Discussion with purported CTOs * Technical debt is subset of organizational debt * Business + technology = "our business"

Community Comment: Part 27 - Assessment is needed when deciding between transactional and analytical modeling

* Transactional versus analytical modeling * Discussion with purported staff data engineer * Making broad generalizations doesn't help * Quoting $ savings in vacuum not advised

Community Comment: Part 26 - With respect to data normalization, performance is key

* Normalized versus denormalized data * Discussion with purported training firm CEO * Performance is important consideration * More factors involved than just record count