Next month, the Open Software Foundation takes its first practical step towards implementing its Level One operating environment standard. The result will either be a fierce technical and mar keting battle, or an irrelevant non-event. A three-day pitched battle next month between rival suppliers of graphical user interfaces will be used to determine the features […]
Next month, the Open Software Foundation takes its first practical step towards implementing its Level One operating environment standard. The result will either be a fierce technical and mar keting battle, or an irrelevant non-event.
A three-day pitched battle next month between rival suppliers of graphical user interfaces will be used to determine the features of the Open Software Foundation’s standard user interface. The Foundation wants suppliers to make their pitch during a conference in Boston starting September 25, which will be open to Foundation members and non-members alike. By the end of the year, the Open Software Foundation hopes to have distilled from the available products an acceptable, achievable specification which may take features from several different products, and can be implemented for the Foundation Level 1 product, due in the first quarter 1990. Tired of waiting We are sick and tired of waiting for standards, said the Foundation’s director of European Operations Henning Oldenburg. We want to force the issue. This radical departure from the Foundation’s previous stance as solely an implementor, rather than a specifier, of standards, underlines the need for the Software Foundation to be seen to come up with product in a hurry – not least because the opposition, AT&T and Sun Microsystems, have a considerable head start in developing Unix System V, Release 4.0. The conference, during which selected suppliers will resent their products in any way they choose, will be repeated in Brussels on October 6 to 8. By September 16, would be contenders have to submit a five-page document outlining their product, from which a shortlist of those which implement the greatest number of desired features will be selected as candidates for presentation at the conference. By the end of the conference, Foundation staff should have a vast range of likes and dislikes – votes, effectively – from attendees; and remember, the conference is open to both members and non-members. The most widely liked features, picked out from attendees’ votes, will be combined into a specification of the ideal Open Software user interface (we know we’re not going to get it, comments Oldenburg). The Open Software Foundation then plans to take the specification back to the shortlisted companies and compare their ideal with the existing products, discuss whether it is possible, and whether the companies are willing, to add the Foundation features to their software, and get them to say how much it will cost. The Foundation should have the responses by the end of November, and will then hold another conference to decide what is achievable and desirable based on both technical considerations and business factors such as the cost and timescales of development. The aim is that by the end of the year, the Foundation should have not only a final specification, complete with clear description of the applications programming interface, of what is achievable for Level 1 – and probably what will come with succeeding Levels – but also should be able to place contracts for the development of the product. According to Oldenburg, the Foundation will give the reasons for its selection of the winner and for the rejection of the others. I believe that the user doesn’t care if the user interface is slightly better or worse, he said. What matters is that there is a standard. Nevertheless, he freely admitted that the Foundation approach of combining features from different products to form a more or less instant standard may may turn out to be over-idealistic. In particular, he conceded that the desperately tight timescales may well overrun – but as he pointed out, setting an early deadline at least pushes the process harder towards a conclusion. The list of prerequisites for user interface technology may ensure that some products that would otherwise be prime candidates are not submitted at all: besides the obvious requirements that the software be X-Window-based, written in C, Unix-compatible and hardware independent, there have to be no licensin
g restrictions apart from royalty considerations – companies have to make the source code available to any Foundation customer and it has to be commercially available by the end of the first quarter 1989. Because the selection of the base technology is announced at the same time as the specificat-ion is published, Oldenburg hopes to be able to encourage the companies that were not selected to change their products to make them compatible with the specification (they all have an equal time during which to implement the changes). He also hopes that applications developers will start writing software to meet the spec so that by the time the user interface is shipped, there should be applications software available. These are high hopes indeed, since the look and feel provided by graphical user interfaces is the next method that many manufacturers hope to use to entice users into the proprietary fold – and an indication of the importance of the issue is the number of lawsuits that are currently flying back and forth between major suppliers. Cucumber sandwiches If the user interface selection process proves reasonably successful, the Foundation hopes to use a similar mechanism to select other technologies, although as Foundation membership grows, future conferences will probably be restricted to members. The next technology in line is database management systems. However, as Oldenburg admits, with Foundation membership standing today at a mere 25-odd, the decision to open the first conference to the world at large was essential if the new body is to have any hope of achieving a market consensus in the selection of the user interface technology. The Boston conference should be an acid test of industry and user opinion; if the Open Software Foundation is viewed as sufficiently relevant, the three-day conference could be the scene of a ferocious technical and marketing battle between competing vendors. If, on the other hand, the conference is characterised by restraint, a lack of controversy, cups of tea and cucumber sandwiches, then the credibility of the Open Foundation as a source of vital technology could be seriously jeopardised.