JCE Software, 5C1: Editorial

Will Your Software Run on My Kind of Computer?

John W. Moore
University of Wisconsin-Madison, Madison, WI 53706-1396

Note:
This issue is out of print.

Our subscribers have often assumed that whatever software we published for one type of computer would also be published for the other hardware types that we support. In general this is not the case, largely because software authors do not write for several types of computers at once--it is more than enough for an author to create and debug a program for a single computer platform, much less do the same for two or three. The knowledge of hardware and software needed for single-platform development is more than great enough to ensure that very few of us can develop even for a single kind of computer. Those who do more should be warmly thanked for service above and beyond the call of duty.
I have sometimes been asked why the editorial staff does not address the cross-platform compatibility problem. In many cases we would like to, and in a few we have, but editing and publishing new issues of JCE: Software and supporting users of existing issues provides more than enough to do. Moreover, each year the number of issues previously published and the support and help we need to provide to users of those issues increases. When you consider that there are several special issues (video laserdiscs, computerized index to the Journal of Chemical Education, etc.) in addition to the six regular issues that we produce each year, our editorial staff of three (full-time equivalent) is spread pretty thin. Therefore it is seldom that we can rewrite programs so that they will work on other types of computers from the one the original author wrote for.
Shortly after Project SERAPHIM began collecting and distributing software, several SERAPHIM Fellows and a number of student programmers translated most of the Apple II programs that had been submitted to the project so that they would run on IBM PCs. This was a major cross-platform development project and required several years to complete. It assured that if a teacher was using an Apple II and suddenly was faced with a new computer room full of IBM hardware (or vice versa), SERAPHIM would be able to supply that teacher with roughly equivalent versions of many of the programs it had available.
Soon thereafter we tried translating some programs to the Macintosh environment, but this was much less successful. The translated programs ran on the Mac, but they failed to take advantage of its many innovative features that every Macintosh user was justified in expecting a Macintosh program to have. Programming for a graphical user interface such as the Mac's was especially difficult in the early years, because no programming tools had been developed to ease the task. SERAPHIM eventually concluded that simple translations did not look right on the Mac, and writing new versions of the programs to make effective use of the graphical user interface would be too time consuming for the resources we had available. Hence SERAPHIM had far fewer Macintosh programs than it had for Apple II or IBM PC.
Since those early Macintosh years, many more tools have become available to help programmers make effective use of a graphical user interface. Some of these (or their functional equivalents) are even available for two or more kinds of computers. Now that Windows and other graphical user interfaces are becoming available on IBM and compatible hardware, we can expect more and more cross-platform tools to become available, because they make it possible for programmers to address a larger total market of potential purchasers. As the capabilities and type of interface become more and more alike across platforms, developing versions of programs that are relatively independent of the hardware on which they were initially designed will become easier and easier.
This year JCE: Software has published two major programs that run on both IBM and Macintosh hardware. One of these is the computerized Index to the Journal of Chemical Education, which appeared in Special Issues SP-5M and SP-5W. The index was created and programmed by Paul Schatz using HyperCard on the Macintosh. The IBM version was translated into ToolBook under Windows by Jerrold Jacobsen. Since most of HyperCard's capabilities are also available in ToolBook, it was relatively easy to make an IBM version that behaves almost exactly the same as the Mac version. Nevertheless, there are a few differences, and it was not possible to use exactly the same documentation for both.
The second cross-platform example is AnswerSheets, by Dick Cornelius, which appears in this issue. Dick began his work on a Mac, but did much of the programming on an IBM under Windows, because it happened that faster IBM hardware was available to him. IBM/Windows AnswerSheets have already been published in JCE: Software, Volume 1D, Number 1. Both versions rely on Microsoft Excel as a programming language. Since Excel is available in both Mac and Windows environments, AnswerSheets created in the Excel macro language can run in both environments. Again, however, there is not exact compatibility. Some things had to be changed because of differences between the two environments, and it appears that our original hope for compatibility across versions of Excel in each environment is not likely to be realized. Nevertheless, Dick has succeeded in creating a series of programs that are close enough to being compatible in two operating environments so that it is possible to publish them for both Mac and Windows.
Things are not yet as simple as we would like them to be, but I am certain that they will become more straightforward in time. Cross-platform development goes on constantly in the commercial software world, and consequently I expect it to become more commonplace in instructional computing as well. The Windows and Macintosh platforms appear to be rapidly moving closer and closer together, and I expect that in the not too distant future there will be little or no distinction between them. JCE: Software hopes to be at the forefront of instructional software development for both, and it appears that making excellent, peer-reviewed software available in versions for both platforms will become easier and easier. That's good news for us, for authors such as Dick Cornelius who want their software distributed as widely as possible, and for chemistry teachers who are eager to use the best new software no matter what kinds of computers are available to them and their students.

First Published: November 1993
Citation: Moore, J. W. Will Your Software Run on My Kind of Computer? J. Chem. Educ. Software 5C1
Keywords:

News | Issues | CD-ROM / Video | Find It! | Technical Support | For Authors

JCE Online | Journal | Software | Internet | Happenings | About JCE | Contact JCE

Last Updated: April 26, 2001
Created: December 3, 1996 Created by: J. L. Holmes
Comments to: jceonline@chem.wisc.edu

© 1997 Division of Chemical Education, Inc., American Chemical Society. All rights reserved.



Will Your Software Run on My Kind of Computer?
John W. Moore University of Wisconsin-Madison, Madison, WI 53706-1396

Note:	This issue is out of print.

Our subscribers have often assumed that whatever software we published for one type of computer would also be published for the other hardware types that we support. In general this is not the case, largely because software authors do not write for several types of computers at once--it is more than enough for an author to create and debug a program for a single computer platform, much less do the same for two or three. The knowledge of hardware and software needed for single-platform development is more than great enough to ensure that very few of us can develop even for a single kind of computer. Those who do more should be warmly thanked for service above and beyond the call of duty. I have sometimes been asked why the editorial staff does not address the cross-platform compatibility problem. In many cases we would like to, and in a few we have, but editing and publishing new issues of JCE: Software and supporting users of existing issues provides more than enough to do. Moreover, each year the number of issues previously published and the support and help we need to provide to users of those issues increases. When you consider that there are several special issues (video laserdiscs, computerized index to the Journal of Chemical Education, etc.) in addition to the six regular issues that we produce each year, our editorial staff of three (full-time equivalent) is spread pretty thin. Therefore it is seldom that we can rewrite programs so that they will work on other types of computers from the one the original author wrote for. Shortly after Project SERAPHIM began collecting and distributing software, several SERAPHIM Fellows and a number of student programmers translated most of the Apple II programs that had been submitted to the project so that they would run on IBM PCs. This was a major cross-platform development project and required several years to complete. It assured that if a teacher was using an Apple II and suddenly was faced with a new computer room full of IBM hardware (or vice versa), SERAPHIM would be able to supply that teacher with roughly equivalent versions of many of the programs it had available. Soon thereafter we tried translating some programs to the Macintosh environment, but this was much less successful. The translated programs ran on the Mac, but they failed to take advantage of its many innovative features that every Macintosh user was justified in expecting a Macintosh program to have. Programming for a graphical user interface such as the Mac's was especially difficult in the early years, because no programming tools had been developed to ease the task. SERAPHIM eventually concluded that simple translations did not look right on the Mac, and writing new versions of the programs to make effective use of the graphical user interface would be too time consuming for the resources we had available. Hence SERAPHIM had far fewer Macintosh programs than it had for Apple II or IBM PC. Since those early Macintosh years, many more tools have become available to help programmers make effective use of a graphical user interface. Some of these (or their functional equivalents) are even available for two or more kinds of computers. Now that Windows and other graphical user interfaces are becoming available on IBM and compatible hardware, we can expect more and more cross-platform tools to become available, because they make it possible for programmers to address a larger total market of potential purchasers. As the capabilities and type of interface become more and more alike across platforms, developing versions of programs that are relatively independent of the hardware on which they were initially designed will become easier and easier. This year JCE: Software has published two major programs that run on both IBM and Macintosh hardware. One of these is the computerized Index to the Journal of Chemical Education, which appeared in Special Issues SP-5M and SP-5W. The index was created and programmed by Paul Schatz using HyperCard on the Macintosh. The IBM version was translated into ToolBook under Windows by Jerrold Jacobsen. Since most of HyperCard's capabilities are also available in ToolBook, it was relatively easy to make an IBM version that behaves almost exactly the same as the Mac version. Nevertheless, there are a few differences, and it was not possible to use exactly the same documentation for both. The second cross-platform example is AnswerSheets, by Dick Cornelius, which appears in this issue. Dick began his work on a Mac, but did much of the programming on an IBM under Windows, because it happened that faster IBM hardware was available to him. IBM/Windows AnswerSheets have already been published in JCE: Software, Volume 1D, Number 1. Both versions rely on Microsoft Excel as a programming language. Since Excel is available in both Mac and Windows environments, AnswerSheets created in the Excel macro language can run in both environments. Again, however, there is not exact compatibility. Some things had to be changed because of differences between the two environments, and it appears that our original hope for compatibility across versions of Excel in each environment is not likely to be realized. Nevertheless, Dick has succeeded in creating a series of programs that are close enough to being compatible in two operating environments so that it is possible to publish them for both Mac and Windows. Things are not yet as simple as we would like them to be, but I am certain that they will become more straightforward in time. Cross-platform development goes on constantly in the commercial software world, and consequently I expect it to become more commonplace in instructional computing as well. The Windows and Macintosh platforms appear to be rapidly moving closer and closer together, and I expect that in the not too distant future there will be little or no distinction between them. JCE: Software hopes to be at the forefront of instructional software development for both, and it appears that making excellent, peer-reviewed software available in versions for both platforms will become easier and easier. That's good news for us, for authors such as Dick Cornelius who want their software distributed as widely as possible, and for chemistry teachers who are eager to use the best new software no matter what kinds of computers are available to them and their students.


News \| Issues \| CD-ROM / Video \| Find It! \| Technical Support \| For Authors
JCE Online \| Journal \| Software \| Internet \| Happenings \| About JCE \| Contact JCE

Last Updated: April 26, 2001 Created: December 3, 1996	Created by: J. L. Holmes Comments to: jceonline@chem.wisc.edu

© 1997 Division of Chemical Education, Inc., American Chemical Society. All rights reserved.