Friday, December 21, 2018

Updates

2018 has been a busy year. It has been some time since I last posted, so I didn't want the year to end without an update.

We formed a new company (Didgets.io) and started a simple web page for it. I have added a ton of features and enhanced many of the previously implemented ones. We successfully found our first two paying customers so we had some modest income this year. We are currently working on adding team members and looking for some working capital to speed up development. We have a number of potential customers that we are currently working with to get them on board.

As every startup founder does, I have to wear multiple hats. The 'Documentation Hat' has been one I have obviously neglected as other tasks have consumed my time.

To catch up, here is a brief list of a few of the most important changes over the past year:

1) Added a lot of Json support to try and capture some of the NoSQL market. Json files can be used to import tables and we can export values and results into a Json file. Since Json supports arrays for every single value, we have been able to test out our 'three dimensional table' features. Each row/column intersection can have multiple values that can be treated separately.

2) Ported everything to Linux and working on the MacOS version too. Updated the build tools to VS2017 and use Qt 5.11.2 for the browser tool. Can now build in both Visual Studio and Qt Creator.

3) Create indexes of sets of text files or of table columns. These are not your typical RDBMS indexes that are used to speed up queries. They are analytical tools to find and analyze patterns in text.

4) Create catalogs of other systems. We can now create Didgets (with associated tags) without importing the data streams based on files in other systems.

5) Enabled 'drill down' analytics from relational table results. If the user has a result set (from either a 'SELECT *' or a more specific query), they can see all the values represented for each column using the 'show values' option. For example: a query against a customers table might display all customers living in California (e.g. 10,000 rows). Right click on the 'city' column header and select the 'show values' option and it will give you a list of all the cities for those 10,000 customers and how many customers are found in each city. If you double click on one cell withing the result set (e.g. 'San Francisco' for row number 3 in the city column) it will pop up a new result set with every customer in that city (e.g. 2500 rows showing each customer in San Francisco). Continuing this process lets the user drill down to more and more specific criteria.

6) Added a set of formulas so our table results can each work much more like a spreadsheet.

7) Added a bunch of transformations to database tables. The user can now modify the data on a column by column basis. We can do things like uppercase, strip punctuation, trim spaces, truncate, replace or split values. The resulting transformations can be placed in entirely new columns within the table.