In the new digital world where data lives everywhere, data catalogs are essential to your Enterprise Data Strategy. Herewith a graphic highlighting some of the key providers.
Almost a year since last adding something. Clear I got some priorities wrong.
Herewith a simple C++ Cmdline routine called DataTrimmer one can use to trim large and super-large datasets with. Written it specifically with C++ to not have dependencies when running. You should be able to copy the EXE file to any windows machine(client or server) and run it.
Many times IT will extract data in the many millions of rows and normal utilities will not be able to deal with it due to its size.
You simply tell the utility how many rows to include and it will trim the data accordingly.
Early notification of the PowerMapping suite I with some associates are bringing to market. Select Customer and Project feedback very good. Also evolve into a Service called MapTo.
It have the ability to rapidly take non-standard arguably unknown data and rapidly correlate it with one or more Industry models. Accelerators for IAA and Guidewire exist with Acord, eTOM, TIA and others to follow.
Suite consist of a a) Mapper (POWERMAPPER) seeded from an Enterprise Repository allowing Project teams to do data mapping whilst the Project is doing what it is suppose to do. It also consist of the b) Validator (POWERVALIDATOR) that can validate mappings and provide degree of compliance. Lastly a cloud based repository service c) POWERDRIVE create enterprise knowledge ensuring mappings are getting more accurate in time. It also drive next gen visualisations amongst others data lineage and flow.
Suite is showing clear value when new Platforms are introduced with heavy duty legacy conversions typical in Insurance(new Product Admin Platforms). Also been used when driving data exploration when designing service interfaces and data warehouses.
What make YOU see the outline of a solution? Whether you are the builder, planner, designer or even the change agent that helps motivating for the new solution… what makes you see it?
The obvious is that any solution must be done for a reason and therefore you must be clear on what you are solving for. In doing so, the biggest problem I observe is that solution crafter’s are ignoring the 3 to FIVE rule.
The rule simply imply that ANYTHING about the solution you write down or draw should never be less than 3 items or more than 5 items.
OK here is the test; look at your latest Powerpoint deck, your Visio outline or even your emails describing the outline of the problem and/or the solution. If you cannot count the items you have a big problem… you are in no-where land. If the points typically are more than 5 you are fuzzy and cannot see the problem/solution and therefore other people that you are communicating with will even see it less. If however, you consistently describe the problem, solution and options respectively in 3-5 points you are seeing it. If you consistently do it in 3 you are on the money.
Current thinking dictates Data Models to be re-presented as ERD’s or UML structures when looking at structured data. Herewith something different when one wants to find patterns within the Data. Example is the IBM IAA BDM Model shown within a Cluster Circle Packing visualisation.
Labels is not shown to respect IBM copyright. Example was specifically chosen to show how it would work across a well matured data model with about 2000 entities and over 1200 attributes in support.
Recently I went through the AppStore thinking it may be quite easy to enter and manage it from a tablet. Though most are visually quite appealing they almost all fail when one want to edit time records, add additional properties or even report slightly different to what the App allows.
This led me back to faithful pivots with simple time-entries in Excel. Herewith sanitised version I use myself, just remember to refresh Pivot for latest views.