The likeliest scenario is that the Canonical repository will limp along, under-used and unnoticed for a few years before someone takes pity and puts it out of its misery.
Last week, Canonical, the commercial face of the Ubuntu GNU/Linux distribution, announced that it would be using its Partners repository to sell proprietary applications like Parallels Workstation. You can see the reasoning: Ubuntu's Debian technology already has the infrastructure for on-demand downloads and software installation, so why not monetize it?
But, if past incarnations of the idea are any indication, then the results are likely to be disappointing at best. For one thing, neither the free software community nor the software vendors care for the idea, so there's little market for it. For another, with the recent maturity of many pieces of free software, how many Ubuntu users will be so insistent on a brand name that they'll pay for functionality that they can get for free?
So far as I know, the idea of a commercial repository was first raised by Ximian back in the dot-com bubble with Red Carpet. At the time, the idea made sense if you squinted a little. Many free productivity apps were light on features, and hopeful startups like Chilliware were rushing to fill the gap with proprietary applications. However, Ximian was acquired by Novell before it could experiment with the idea very much, and the sudden collapse of the bubble took most of the potential vendors for Red Carpet with it.
The next major incarnation was Linspire's Click-N-Run (CNR), which still exists. Linspire claims that CNR contains over 20,000 applications. But perhaps a better indication of the service's success may be that, in the last year, Linspire has offered basic access for free, and extended support to include non-Debian based distributions.
Certainly, the service seems to have done little to move Linspire from the second or third tier of distros. Nor do any major distributions offer to connect users to CNR, though at one point Canonical was considering partnering with Linspire in CNR rather than creating its own commercial repository.
The latest version of the idea is Windows Marketplace. Few, if any Windows users are aware of it, much less using it, despite the fact that the number of applications it offers has grown steadily since Vista was released. For all the marketing logic behind the idea, it seems that even association with the world's most widely used operating system cannot make a commercial repository more than an indifferent success. With this history, you have to wonder why Canonical has bothered reviving the idea, especially when conditions for success seem poorer than ever.
The dangers of a backlash
One problem with the idea is the danger of a backlash within the community. True, only a minority of GNU/Linux users refuse to use any proprietary software whatsoever. However, most feel uncomfortable doing so. Moreover, to judge from the reaction when Fedora made proprietary sound codecs available through the desktop, many object to having access to proprietary software made too easy.
Although this attitude is hypocritical and is unlikely to be shared by the new users that Ubuntu is increasingly targeting, it is still an issue that the Partners repository has to face. In fact, if you read the news about the repository carefully, you can sense Canonical bracing for a negative reaction, as its representatives stress that use of the repository is voluntary and emphasizes convenience and choice. They mention that both free and proprietary applications will be available through it.
So far, the backlash hasn't happened. Yet neither has any great enthusiasm. Judging by the Ubuntu forums, the most common reaction has been mild curiosity, followed by efforts to resolve individual technical problems.
While the possibility of a later backlash can never be discounted, one reason that it may fail to materialize is simple lack of interest in the idea. Today, major software vendors and hardware manufacturers routinely offer free readers, media players and device drivers for GNU/Linux, yet otherwise no more proprietary software is available for the operating system than at the height of the dot-com boom.
Some software, such as Anime Studio Pro, has sprouted GNU/Linux versions because of pressure from users; others, such as Nero, have offered half-hearted versions of their Windows apps -- and that's all. Despite frequent requests from users, major proprietary vendors like Adobe seem no more interested in developing for the platform now than they were seven years ago.
In fact, if you look at the CNR warehouse, what is striking is the scarcity of proprietary products. Not surprisingly, you see more in the Windows Marketplace, yet, even there, few of the major vendors other than Microsoft have any products listed. Instead, what you see are mostly small-timers and shareware developers of specialized products -- and these are exactly the kind of applications that free software has always excelled at producing.
Look, for example, at Parallels Workstation, the Partners repository's first proprietary offering. It's priced at $49.99, which is more than a reasonable price for virtualization software. Yet VirtualBox does the same thing and is free for the download. For that matter, you can download VMware Player at no cost, even if its not philosophically free. With this competition, how many would want to pay to download Parallels Workstation?
Even if major vendors were to release GNU/Linux versions of their products, the enthusiasm would likely be muted. Seven years ago, a GNU/Linux version of Microsoft Office or PhotoShop would have been big news, and cornered the market. But, in the interval, programs like OpenOffice.org or the GIMP have filled the gap created by the proprietary vendor's reluctance with applications of roughly equivalent functionality.
Users -- especially unimaginative ones -- may carp about the lack of one particular feature or the other, but in the major productivity areas, free solutions are already available. Even those who have no political or philosophical objection to proprietary software are going to hesitate about paying for functionality that they can get for free.
A possible exception to this situation may be high-end applications such as the GNU/Linux versions of the 3-D modeling software Maya or the Oracle database. The trouble is, these apps are beyond the budgets of home users, and no well-run business is going to buy software without a procurement process that negates the speed and convenience of the Partners repository.
Perhaps this line of reasoning is what keeps so many proprietary vendors away from similar download services. At any rate, the service is a new distribution model, and mostly untried, both by vendors and by users coming from Windows. Both seem more comfortable with the retail store than the download service, and are likely to remain so for a few years.
A limited niche
A download service might find a temporary niche in offering software for which no free equivalent exists. For instance, despite recent improvements in apps like Kooka and Tesseract, someone who regularly needed to convert scanned text to a usable format might welcome a GNU/Linux version of OmniPage. The trouble is, given the speed with which free software is developing, such a market would be temporary, lasting a year or two at most. A service specializing in these niches would continually lose out to maturing free software, with no prospect of replacement products.
Similarly, users migrating from Windows might become temporary customers because they retain the assumption that proprietary software is superior. But, if they remained with Ubuntu, they would be lost as customers as they learned to trust free software.
Canonical has been considering such a service for over a year, but you have to wonder why. The users who are comfortable with this type of service have no need of it, while those home and business users who might find it useful or reassuring are not comfortable with it -- even if they could find products worth buying.
Perhaps, as Canonical claims, the service is a response to requests from partners. Yet it is so poorly considered that you might wonder how familiar its developers were with its market. Or, had so much time and effort been spent on the project that nobody had the courage to cut the company's losses before it was completed?
If any company could make the idea pay, it would be an innovative one like Canonical. However, the likeliest scenario is that the Partners repository will limp along under-used and unnoticed for four or five years before someone takes pity and puts it out of its misery.