Interoperability

All modern software needs to be able to act on shared data with third party applications. This is especially true in the situations in which portalBase is commonly used, for example where it is a rapid development environment for creating applications to supplement legacy or hard-to-alter core software. You rarely get full value out of data if it’s in a ‘silo’ that no one else can access (see http://en.wikipedia.org/wiki/Information_silo). 

Some examples of interoperability are

  • allowing an e-commerce site to be updated with stock availability and prices from from a central stock control / ERP system
  • allowing desktop publishing software access to the marketing dept.’s image/document resource library
  • Electronic Data Interchance (EDI) – allow large amounts of orders to be processed automatically in a supply chain

Today’s release of portalBase 2.2.4 adds more options to make interoperability easier. You’ve always been able to import data from third party systems, via spreadsheet and also to export in the same way. However, you can now update existing data using a new enhanced import process. This allows a round-trip of export/modify/import or regular updates from external sources, e.g. updating of product prices in an invoicing system or contact details in a CRM system.

Here’s the first step of the new import wizard.

import_wizard

Remember, for completely automated data transfer, clients can also request direct database access from other software to their underlying portalBase database with an industry standard interface (ODBC / JDBC to postgreSQL, for technical readers)

Other improvements in this release

  • Calculations have been improved – users can now create one calculation that references another
  • Error messages have been made clearer both for calculation creation and the import process
  • the iPhone version of portalBase is stable – more news about this shortly!

Learn how to filter

The ability to filter data is one of the things that makes portalBase really powerful. Searching and filtering is quite easy but many users tend to stick with basic filters. This release of version 2.2 adds some more options so now’s a good time to write a quick tutorial by example. For those who want to learn, it couldn’t be simpler, read on!

Basic filtering

To find text starting with a word or letters, just type. e.g. is searching an address book:

Type john to find John Smith, John Jones and John Barleycorn

Note filtering is case insensitive

To find text anywhere in a field, use the star character as a ‘wildcard’:

Type *smith to find John Smith, Jane Smith and Tobermory Smith

Some more options

help1If you forget these options, don’t worry they’re in the help built into portalBase. Move your mouse over the question mark at the left of the filter boxes to get a quick reminder.

All of the syntax below works any type of data, be it text, numeric or date. Some examples for numbers:

>400 returns all those greater than or equal to 400

< 600 returns all those less than 600

New:
>400 and <600 returns all those in the range

In a text field, e.g. a list of company names,

*Ltd or *Limited will return all limited companies

!google will return all companies that are NOT Google, or at least don’t start with the word Google

New:
!google and !yahoo will return everything except those two companies

? will return all blank entries, i.e. where the company name isn’t filled in at all

Conversly, !? (‘not blank’) will return all those that are

New: dates and times

Date filtering is a new addition that deserves a special mention because you can filter in lots of ways. Here are just a few examples:

today will find any times that are sometime today

next week any date falling in the next week

two months ago

2009

Jan 5th 2010

<ten years ago anything more than ten years old

>today anything in the future

>01 Jan 2009 and <31 March 2009 dates in the first quarter

*Jan dates in January from any year – useful for comparing year on year

etc.

Try your own! If you find any that don’t give the results you expect, please contact oliver@gtportalbase.com

Credits: this work is partly based on the Ruby Chronic library and the Java port by  mDimension Technology

Tag clouds from content in portalBase 2.1

Recently some work with a UK Sector Skills Council has involved investigating tag clouds for visualising database content.

We thought what better way to test out some ideas than to rapidly prototype the database structure with portalBase and plug in a facility for visualising a cloud of words from data in it? We quickly did this and found some useful results, but it also turns out that tag clouds are a great way of summarising lots of different types of data, so the facility is now generally available, tweaked and honed in GTpB version 2.1!

To see a tag cloud of your own data, simply go to the summary tab of any report. The cloud automatically appears as part of the summary and will change as you filter data. It’s really useful for bringing new people up to speed with systems as it gives a quick overview of what the data’s all about.

Here are some examples –

A cloud from our own projects database:

projects-tag-cloud

Words from the issue log for a web application software development:

issue-log-tag-cloud

One of the reports from a food sales invoicing system:

food-tag-cloud

Multi-browser compatible!

GT portalBase is now compatible with Firefox, Safari, Chrome and Internet Explorer with the release of version 2.0!

browsers

In other news, field names and descriptions can be edited by clicking on them in the ‘fields’ tab of a table. Along with the  drag and drop field re-ordering from the previous release, this makes schema editing truly agile.

Version 1.9.9 deals with large data sets

This release addresses a number of items to improve input and output of large amounts of data, both from the schema design and end user points of view

Image:Import.png Importing 

  • Metadata: fields such as creator and creation date are set for every imported record
  • Bugfixes to the import routine – portalBase should deal with Excel-generated CSV files better, with errors less common
Image:Fields.png Fields 

  • Fields can be re-ordered very quickly now, just by by dragging them, in the ‘fields’ tab of pane 3. This vastly speeds up modifications to large tables
  • Separators can be added between fields to section off different parts of an input form, e.g. to clarify stages in a process
Image:Print.png  Document generation 

  • The print facility now has the ability to include full detail for every record printed in the currently selected report, one record per page. This allows easy generation of e.g. detailed specification documents from data in portalBase

Other improvements

In this and recent minor releases, a number of other improvements and bugfixes have been included

  • Field sorting is improved with visual feedback
  • Usage statistics: administrators can easily see on a per-report or per-table basis who’s using each module. De-cluttering obsolete reports is easier
  • Report generation is a lot quicker – fields and joins are automatically added to a report on creation
  • Addresses: geographical location is displayed next to phone numbers. Addresses are auto-recognised and labelled with map icons
  • The ‘link’ button now works more simply, launching an email with a link to the current portalBase record
   

The client side – version 1.9 released

Over the past few months, many minor changes have been released to address particular requirements, but this major release draws together a number of features. With the database and server components solid and high performance, this release focusses on the user interface. We’ve had UI specialists working on specific areas and the results are described below.

 

  Image:Pane_1.png  

Sections

Pane 1 can now be arranged into sections. For example, you may like to arrange modules into Marketing/Sales/R&D/Accounts, Plan/Do/Check/Act or whatever suits your organisation. 

Speed

We’ve targeted slower internet connections in this release. In particular, pane 3 requires much less bandwidth to download, as only the current tab is loaded when selecting a new record. Large reports now see a 4x speed improvement when loading into pane 2 due to the use of data compression during data transfer from the server to the browser. 

Foundations for the Future

Behind the scenes, the whole of the portalBase interface has also been rewritten to be more browser compatible. This work isn’t fully complete but it does mean that our next release will be compatible with, amongst other browsers, Internet Explorer as well as Firefox.

 

version 1.8.5 brings iPhone platform and intelligent report creation

New features

  • Simple joins – life is now simpler for report creators: the system guides you when collating data from multiple sources into one report. Simple, context sensitive options are given e.g. when editing a contacts report, ‘join contacts to addresses’ may be offered. A knowledge of SQL principles is no longer necessary to create most joins, however it’s still possible to create complex joins, e.g. joining on calculations, by people who do have the necessary technical skills
  • Improved logging: audit trails now include details such as filters applied to a report and the actual data entered while editing a record

Bugfixes

  • Minor bugfixes, including fixing one case in which importing data from a spreadsheet would fail
Image:PortalBase-on-iPod.png One more thing 

We’re working on a mobile of version of portalBase, specifically tailored for devices such as the iPhone, iPod Touch or newer Symbian based mobile phones. It’s currently early in the development process and has facilities such as

 

  • module viewing
  • report viewing and filtering as-you-type
  • summary report viewing
  • data editing