Difference between revisions of "User talk:Bucheron"
Jump to navigation
Jump to search
m |
m |
||
| Line 1: | Line 1: | ||
Well, this is the place where you can chat with me !! | Well, this is the place where you can chat with me !! | ||
| − | My vision of pdb | + | == My vision of pdb == |
* a central RDBMS managing one or more databases | * a central RDBMS managing one or more databases | ||
| − | + | ::- databases with strong long-term definitions of object classes | |
| − | + | ::- with event data synchronization capability on regional "slave" servers | |
* Dedicated projects separated by "domain of interest", each implementing services or micro-services on https servers | * Dedicated projects separated by "domain of interest", each implementing services or micro-services on https servers | ||
(eg interaction with the public, interaction with "manufacturers", synchronization with external data, media management, authorization management, etc.) | (eg interaction with the public, interaction with "manufacturers", synchronization with external data, media management, authorization management, etc.) | ||
| Line 11: | Line 11: | ||
| − | My doubts | + | == My doubts == |
* Is it a good idea to split the "attributes" of an object into two parts : | * Is it a good idea to split the "attributes" of an object into two parts : | ||
| − | + | ::- id (PK), uniqueName ... in SQL columns | |
| − | + | ::- attributes "nullable" in one or more JSONB columns | |
* how to manage object "variants" | * how to manage object "variants" | ||
* how to define "family", "category", "type", "group", "product" ... for an efficient hierarchy | * how to define "family", "category", "type", "group", "product" ... for an efficient hierarchy | ||
| Line 22: | Line 22: | ||
| − | Softwares I use | + | == Softwares I use == |
* Proxmox | * Proxmox | ||
* Debian | * Debian | ||
Revision as of 13:29, 29 March 2019
Well, this is the place where you can chat with me !!
My vision of pdb
- a central RDBMS managing one or more databases
- - databases with strong long-term definitions of object classes
- - with event data synchronization capability on regional "slave" servers
- Dedicated projects separated by "domain of interest", each implementing services or micro-services on https servers
(eg interaction with the public, interaction with "manufacturers", synchronization with external data, media management, authorization management, etc.)
- authentication with a key in HTTP header (no session)
- role-based authorizations
My doubts
- Is it a good idea to split the "attributes" of an object into two parts :
- - id (PK), uniqueName ... in SQL columns
- - attributes "nullable" in one or more JSONB columns
- how to manage object "variants"
- how to define "family", "category", "type", "group", "product" ... for an efficient hierarchy
- is it possible utilize a world-wide unique product-id
- where to place internationalization (database or file-system)
- do we need to manage "distributors"
Softwares I use
- Proxmox
- Debian
- Nginx
- PostgreSQL (PostGIS)
- Java
- Vert.x
- Javascript
- Gitlab
- Redmine
- Mattermost
- Zimbra
