Microsoft Access

Project Omega

Microsoft's first attempt to sell a relational database product was during the mid-1980s, when Microsoft obtained the license to sell R:Base.^[11] In the late 1980s Microsoft developed its own solution codenamed Omega.^[12] It was confirmed in 1988 that a database product for Windows and OS/2 was in development.^[13][14] It was going to include the "EB" Embedded Basic language,^[12] which was going to be the language for writing macros in all Microsoft applications,^[15] but the unification of macro languages did not happen until the introduction of Visual Basic for Applications (VBA). Omega was also expected to provide a front end to the Microsoft SQL Server.^[16] The application was very resource-intensive, and there were reports that it was working slowly on the 386 processors that were available at the time.^[17] It was scheduled to be released in the 1st quarter of 1990,^[18] but in 1989 the development of the product was reset^[11][19] and it was rescheduled to be delivered no sooner than in January 1991.^[20] Parts of the project were later used for other Microsoft projects: Cirrus (codename for Access) and Thunder (codename for Visual Basic, where the Embedded Basic engine was used).^[11][12] After Access's premiere, the Omega project was demonstrated in 1992 to several journalists and included features that were not available in Access.^[21]

Project Cirrus

After the Omega project was scrapped, some of its developers were assigned to the Cirrus project (most were assigned to the team which created Visual Basic).^[11] Its goal was to create a competitor for applications like Paradox or dBase that would work on Windows.^[22] After Microsoft acquired FoxPro, there were rumors that the Microsoft project might get replaced with it,^[23] but the company decided to develop them in parallel. It was assumed that the project would make use of Extensible Storage Engine (Jet Blue)^[24] but, in the end, only support for Jet Database Engine (Jet Red) was provided. The project used some of the code from both the Omega project and a pre-release version of Visual Basic.^[12] In July 1992, betas of Cirrus shipped to developers^[25] and the name Access became the official name of the product.^[26] "Access" was originally used for an older terminal emulation program from Microsoft. Years after the program was abandoned, they decided to reuse the name here.^[27]

Timeline

Microsoft released Access version 1.0 on November 16, 1992,^[1] announcing its immediate availability at the fall Comdex trade show, and an Access 1.1 release in May 1993 to improve compatibility with other Microsoft products and to include the Access Basic programming language.

With Access v2.0, Microsoft specified the minimum hardware requirements to be Microsoft Windows v3.1 with 4 MB of RAM required, 6 MB RAM recommended; 8 MB of available hard disk space required, 14 MB hard disk space recommended. The product shipped on seven 1.44 MB diskettes. The manual shows a 1994 copyright date.

As a part of the Microsoft Office 4.3 Professional with Book Shelf, Microsoft Access 2.0 was included with first sample databases "NorthWind Trader" which covered every possible aspect of programming a database. The Northwind Traders sample first introduced the Main Switchboard features new to Access 2.0 for 1994.

With Office 95, Microsoft Access 7.0 (a.k.a. "Access 95") became part of the Microsoft Office Professional Suite, joining Microsoft Excel, Word, and PowerPoint and transitioning from Access Basic to VBA. Since then, Microsoft has released new versions of Microsoft Access with each release of Microsoft Office.^[28]

Versions 3.0 and 3.5 of Jet Database Engine (used by Access 7.0 and the later-released Access 97, respectively) had a critical issue which made these versions of Access unusable on a computer with more than 1 GB of memory.^[29] While Microsoft fixed this problem for Jet 3.5/Access 97 post-release, it never fixed the issue with Jet 3.0/Access 95.

The native Access database format (the Jet MDB Database) has also evolved over the years. Formats include Access 1.0, 1.1, 2.0, 7.0, 97, 2000, 2002, and 2007. The most significant transition was from the Access 97 to the Access 2000 format; which is not backward compatible with earlier versions of Access. As of 2011,^[update] all newer versions of Access support the Access 2000 format. New features were added to the Access 2002 format which can be used by Access 2002, 2003, 2007, and 2010.

Microsoft Access 2000 increased the maximum database size to 2 GB from 1 GB in Access 97.

Microsoft Access 2007 introduced a new database format: ACCDB. It supports links to SharePoint lists and complex data types such as multi-value and attachment fields. These new field types are essentially recordsets in fields and allow the storage of multiple values or files in one field. Microsoft Access 2007 also introduced File Attachment field, which stored data more efficiently than the OLE (Object Linking and Embedding) field.

Microsoft Access 2010 introduced a new version of the ACCDB format supported hosting Access Web services on a SharePoint 2010 server. For the first time, this allowed Access applications to be run without having to install Access on a user's PC and was the first support for Mac users. Any user on the SharePoint site with sufficient rights could use the Access Web service. A copy of Access was still required for the developer to create the Access Web service, and the desktop version of Access remained part of Access 2010. The Access Web services were not the same as the desktop applications. Automation was only through the macro language (not VBA) which Access automatically converted to JavaScript. The data was no longer in an Access database but SharePoint lists. An Access desktop database could link to the SharePoint data, so hybrid applications were possible so that SharePoint users needing basic views and edits could be supported while the more sophisticated, traditional applications could remain in the desktop Access database.

Microsoft Access 2013 offers traditional Access desktop applications plus a significantly updated SharePoint 2013 web service.^[30] The Access Web model in Access 2010 was replaced by a new architecture that stores its data in actual SQL Server databases. Unlike SharePoint lists, this offers true relational database design with referential integrity, scalability, extensibility and performance one would expect from SQL Server.^[31] The database solutions that can be created on SharePoint 2013 offer a modern user interface designed to display multiple levels of relationships that can be viewed and edited, along with resizing for different devices and support for touch. The Access 2013 desktop is similar to Access 2010 but several features were discontinued including support for Access Data Projects (ADPs),^{[clarification needed]} pivot tables, pivot charts, Access data collections, source code control, replication, and other legacy features.^[32] Access desktop database maximum size remained 2 GB (as it has been since the 2000 version).

Microsoft Access is no longer included in the one-time-purchase version of Microsoft Office 2021, but remains within the Microsoft 365 counterpart, as Apps for business and Business Standard editions. Features new to Access 2021 include an extended-precision date/time data type^[33] and dark theme support.^[34]

In addition to using its own database storage file, Microsoft Access also may be used as the 'front-end' of a program while other products act as the 'back-end' tables, such as Microsoft SQL Server and non-Microsoft products such as Oracle and Sybase. Multiple backend sources can be used by a Microsoft Access Jet Database (ACCDB and MDB formats). Similarly, some applications such as Visual Basic, ASP.NET, or Visual Studio .NET will use the Microsoft Access database format for its tables and queries. Microsoft Access may also be part of a more complex solution, where it may be integrated with other technologies such as Microsoft Excel, Microsoft Outlook, Microsoft Word, Microsoft PowerPoint and ActiveX controls.

Access tables support a variety of standard field types, indices, and referential integrity including cascading updates and deletes. Access also includes a query interface, forms to display and enter data, and reports for printing. The underlying Access database, which contains these objects, is multi-user and handles record-locking.

Repetitive tasks can be automated through macros with point and click options. It is also easy to place a database on a network and have multiple users share and update data without overwriting each other's work. Data is locked at the record level which is significantly different from Excel which locks the entire spreadsheet.

There are template databases within the program and for download from Microsoft's website. These options are available upon starting Access and allow users to enhance a database with predefined tables, queries, forms, reports, and macros. Database templates support VBA code, but Microsoft's templates do not include VBA code.

Programmers can create solutions using VBA, which is similar to Visual Basic 6.0 (VB6) and used throughout the Microsoft Office programs such as Excel, Word, Outlook and PowerPoint. Most VB6 code, including the use of Windows API calls, can be used in VBA. Power users and developers can extend basic end-user solutions to a professional solution with advanced automation, data validation, error trapping, and multi-user support.

The number of simultaneous users that can be supported depends on the amount of data, the tasks being performed, level of use, and application design. Generally accepted limits are solutions with 1 GB or less of data (Access supports up to 2 GB) and it performs quite well with 100 or fewer simultaneous connections (255 concurrent users are supported).^[35] This capability is often a good fit for department solutions. If using an Access database solution in a multi-user scenario, the application should be "split". This means that the tables are in one file called the back end (typically stored on a shared network folder) and the application components (forms, reports, queries, code, macros, linked tables) are in another file called the front end. The linked tables in the front end point to the back end file. Each user of the Access application would then receive his or her own copy of the front end file.

Applications that run complex queries or analysis across large datasets would naturally require greater bandwidth and memory. Microsoft Access is designed to scale to support more data and users by linking to multiple Access databases or using a back-end database like Microsoft SQL Server. With the latter design, the amount of data and users can scale to enterprise-level solutions.

Microsoft Access's role in web development prior to version 2010 is limited. User interface features of Access, such as forms and reports, only work in Windows. In versions 2000 through 2003 an Access object type called Data Access Pages created publishable web pages. Data Access Pages are no longer supported. The Jet Database Engine, core to Access, can be accessed through technologies such as ODBC or OLE DB. The data (i.e., tables and queries) can be accessed by web-based applications developed in ASP.NET, PHP, or Java. With the use of Microsoft's Terminal Services and Remote Desktop Application in Windows Server 2008 R2, organizations can host Access applications so they can be run over the web.^[36] This technique does not scale the way a web application would but is appropriate for a limited number of users depending on the configuration of the host.

Access 2010 allows databases to be published to SharePoint 2010 web sites running Access Services. These web-based forms and reports run in any modern web browser. The resulting web forms and reports, when accessed via a web browser, do not require any add-ins or extensions (e.g., ActiveX and Silverlight).

Access 2013 can create web applications directly in SharePoint 2013 sites running Access Services. Access 2013 web solutions store its data in an underlying SQL Server database which is much more scalable and robust than the Access 2010 version which used SharePoint lists to store its data.

Access Services in SharePoint has since been retired.^[37]

A compiled version of an Access database (file extensions .MDE /ACCDE or .ADE; ACCDE only works with Access 2007 or later) can be created to prevent users from accessing the design surfaces to modify module code, forms, and reports. An MDE or ADE file is a Microsoft Access database file with all modules compiled and all editable source code removed. Both the .MDE and .ADE versions of an Access database are used when end-user modifications are not allowed or when the application's source code should be kept confidential.

Microsoft also offers developer extensions for download to help distribute Access 2007 applications, create database templates, and integrate source code control with Microsoft Visual SourceSafe.

Access Services and Web database

ASP.NET web forms can query a Microsoft Access database, retrieve records and display them on the browser.^[38]

SharePoint Server 2010 via Access Services allows for Access 2010 databases to be published to SharePoint, thus enabling multiple users to interact with the database application from any standards-compliant Web browser. Access Web databases published to SharePoint Server can use standard objects such as tables, queries, forms, macros, and reports. Access Services stores those objects in SharePoint.^[39]

Access 2013 offers the ability to publish Access web solutions on SharePoint 2013. Rather than using SharePoint lists as its data source, Access 2013 uses an actual SQL Server database hosted by SharePoint or SQL Azure.^[40] This offers a true relational database with referential integrity, scalability, maintainability, and extensibility compared to the SharePoint views Access 2010 used. The macro language is enhanced to support more sophisticated programming logic and database level automation.^[31]

Import or link sources

Microsoft Access can also import or link directly to data stored in other applications and databases.^[6] Microsoft Office Access 2007 and newer can import from or link to:

• Microsoft Access
• Excel
• SharePoint lists
• Plain text
• XML
• Outlook
• HTML
• dBase (dropped in Access 2013; restored in Access 2016)^[41]
• Paradox (with Access 2007; dropped in Access 2010^[42])
• Lotus 1-2-3 (dropped in Access 2010^[42])
• ODBC-compliant data containers, including:
  • Microsoft SQL Server
  • Oracle
  • MySQL
  • PostgreSQL
  • IBM Lotus Notes
  • IBM i IBM Db2

Split database architecture

Microsoft Access applications can adopt a split-database architecture. The single database can be divided into a separate "back-end" file that contains the data tables (shared on a file server) and a "front-end" (containing the application's objects such as queries, forms, reports, macros, and modules). The "front-end" Access application is distributed to each user's desktop and linked to the shared database. Using this approach, each user has a copy of Microsoft Access (or the runtime version) installed on their machine along with their application database. This reduces network traffic since the application is not retrieved for each use. The "front-end" database can still contain local tables for storing a user's settings or temporary data. This split-database design also allows development of the application independent of the data. One disadvantage is that users may make various changes to their own local copy of the application and this makes it hard to manage version control. When a new version is ready, the front-end database is replaced without impacting the data database. Microsoft Access has two built-in utilities, Database Splitter^[54] and Linked Table Manager, to facilitate this architecture.

Linked tables in Access use absolute paths rather than relative paths, so the development environment either has to have the same path as the production environment or a "dynamic-linker" routine can be written in VBA.

For very large Access databases, this may have performance issues and a SQL backend should be considered in these circumstances. This is less of an issue if the entire database can fit in the PC's RAM since Access caches data and indexes.

Migration to SQL Server

To scale Access applications to enterprise or web solutions, one possible technique involves migrating to Microsoft SQL Server or equivalent server database. A client–server design significantly reduces maintenance and increases security, availability, stability, and transaction logging.

Access 2000 through Access 2010 included a feature called the Upsizing Wizard that allowed users to upgrade their databases to Microsoft SQL Server, an ODBC client–server database. This feature was removed from Access 2013. An additional solution, the SQL Server Migration Assistant for Access (SSMA), continues to be available for free download from Microsoft.^[55]

A variety of upgrading options are available.^[56] After migrating the data and queries to SQL Server, the Access database can be linked to the SQL database, subject to data type conversion limitations:

• "Yes/No" types: In Microsoft Access there are three states for the Yes/No (True/False) data type: empty, no/false (0) and yes/true (−1). The corresponding SQL Server data type is binary, with only two states; the permissible values are 0 and 1.
• Date/Time (Extended): SQL Server's smalldatetime format has precision of 1 minute, minimum date value is 1900-01-01, maximum date value is 2079-06-06; datetime format has precision of 10/3 milliseconds (rounded to increments of .000, .003, or .007 seconds), minimum date value is 1753-01-01, maximum date value is 9999-12-31; datetime2 format has precision of 100 nanoseconds, minimum date value is 0001-01-01, same maximum date value as datetime. Access's Date/Time and Date/Time Extended have higher precisions than smalldatetime and datetime and datetime2, and wider date range than SQL Server's smalldatetime and datetime formats.

Regardless, SQL Server is still the easiest migration. Retrieving data from linked tables is optimized to just the records needed, but this scenario may operate less efficiently than what would otherwise be optimal for SQL Server. For example, in instances where multi-table joins still require copying the whole table across the network.

In previous versions of Access, including Access 2010, databases can also be converted to Access Data Projects (ADP) which are tied directly to one SQL Server database. This feature was removed from Access 2013. ADP's support the ability to directly create and modify SQL Server objects such as tables, views, stored procedures, and SQL Server constraints. The views and stored procedures can significantly reduce the network traffic for multi-table joins. SQL Server supports temporary tables and links to other data sources beyond the single SQL Server database.

Finally, some Access databases are completely replaced by another technology such as ASP.NET or Java once the data is converted. However, any migration may dictate major effort since the Access SQL language is a more powerful superset of standard SQL. Further, Access application procedures, whether VBA and macros, are written at a relatively higher level versus the currently available alternatives that are both robust and comprehensive. The Access macro language, allowing an even higher level of abstraction than VBA, was significantly enhanced in Access 2010 and again in Access 2013.

In many cases, developers build direct web-to-data interfaces using ASP.NET, while keeping major business automation processes, administrative and reporting functions that do not need to be distributed to everyone in Access for information workers to maintain.

While all Access data can migrate to SQL Server directly, some queries cannot migrate successfully. In some situations, you may need to translate VBA functions and user defined functions into T–SQL or .NET functions / procedures. Crosstab queries can be migrated to SQL Server using the PIVOT command.