Linked by Thom Holwerda on Fri 22nd May 2009 13:58 UTC, submitted by shaneco
GNU, GPL, Open Source Keith Curtis worked at Microsoft for 11 years, coding on Windows, Office, and at Microsoft's research department, before leaving the Redmond giant. Call it a revelation, call it giving in to the devil's temptations, but he's now a complete open source and Linux advocate, and in his new book, "After the Software Wars", he explains why open source will prevail against Microsoft's proprietary model.
Permalink for comment 365181
To read all comments associated with this story, please click here.
RE: Comment by moleskine
by karl on Sat 23rd May 2009 19:26 UTC in reply to "Comment by moleskine"
karl
Member since:
2005-07-06

First off I would like to thank you, moleskin and cb_osn, for actually dragging this conversation out of the gutter. Far too often I read something on OSNEWS only to find the discussion drift off into lala land with no end to polemic hogwash.

Although I do not possess a crystal ball which would enable me to foretell the future I do feel there are several factors which will likely lead to a predominance of FOSS software in the future.

Just to hit upon a few factors to clarify: most of what we call propietary software is not the product of larger software corporations. Most of this software is written in-house by firm employees of companies which use software for their business model but which themselves do not sell software. This scenario is starting change in regards to FOSS but I believe there are limits as to how far FOSS will replace proprietary in-house development. Much of what constitutes the business model of many, many, companies is the acquisition, access, and utilization of data. In other words the core of the business is the data and the business in question revolves around how the data is acquired, accessed and utilized-and precisely this "how" constitutes the profitability of the company in question-and ultimately *is* the company, for companies(not all, but some really large number of) are little other than novel ways of acquiring, accessing and utilizing vast sets of data. FOSS can play a significant role in such a scenario-but the identity and raison d'etre of the company lies in it's "own" (ownership, propietary) particular way of dealing with data.


The major software companies came into existence at a time when "data" was just becoming something which one could base companies on. These companies transformed the market and changed what it means to do business. Prior to the rise of companies like Microsoft there were relatively few companies whose core value was data and each of these companies were wholly dependent on the companies which produced computers and the software needed to put these machines to use. In a certain sense Microsoft is also a child of the fabless semiconductor revolution of the 70's: Once it became possible to design chips independent of the actual production of the chips the semiconductor market rocketed and as is the case with Microsoft the ability to produce software without having to actually produce the hardware it would run on allowed the nascent propietary software industry to rise to such towering hights.

Whereas "data" was a somewhat novel notion for the fist 50 years of computer history and an entire industry grew to provide hardware and software for acquiring data, data is now ubiquitous and the emphasis has shifted to managing (accessing, utilizing, rendering etc.) data. The vast majority of programming jobs (no statistical data to back this up) in the past 30 years have revolved around freeing data stored in a particular way(primarily determined by what software was used to acquire it and the software (db) used to store it in order to expose new relations and associations-precisely these new methods of managing the data were the "value add", ie. the profitability.

A large part of the aftermarket proipetary software world consisted of products to translate data from one form into another. This was necessary because of the format of the data( the "how" I spoke of earlier is that which constitutes the format of the data and the value of the data independant of the format approaches 0). The format of the data (ie. how it is acquired, accessed and utilized) is that which renders rather meaningless 0's and 1's into something of value, because the "value" of data is determined by tada, the propietary(ie. exclusive) format. FOSS has made tremendous inroads in this area leaving many programmers looking for a new job because with FOSS the value that data represents is no longer embedded in the format, because format no longer means propietary. Data by itself is worthless, it is the combination of data + data format which is the traditional value proposition.

Where FOSS has made inroads is in the decoupling of this value proposition. In FOSS the data format is still particular(ie. there exists a specific way which constitutes the communication and storage of data)(ie. there is still a moment of "ownership"/"authorship", "propietariness", in FOSS data format) but due to the availability of the code and code sharing there is no longer a need for a propietary aftermarket software industry catering to the needs of data rendered useless due to incompatible formats and protocols.

Microsoft had such "mindshare" because it gave birth to an entire parasitic industry designed to work around the incompatibilities of all the proprietary formats used by the large proprietary software applications, which themselves had a parasitic relationship to the proprietarty OS(Windows).

FOSS lowers the barrier to entry for participating in directly in the management of data. The world at large is less dependent on proprietary applications and proprietary OS's. The value proposition of the propietary software world, ie. the combining of data and data format in such a way as to be incompatible not only with other software, but other ways of acquiring, accessing and utilizing data is under a sustained attack from FOSS and in all likelihood will never regain the market it once had.

Ironically there seems to be a significant relationship between "ease of use" and exterior compatibility, ie. the hallmark of many of the great of the proprietary applications and OS's is there so-called "ease of use", but issues of incompatibility were equally hallmarks of such software. This need not be so, but historically this has been the case, at least apparently. Of course much of what counts as "ease of use" lied therein that there existed exactly one way, one application, of acquiring, accessing and utilizing data, and this one way, one application for one specific task was itself the hallmark of proprietary applications(if you want this, you need to purchase this). FOSS, by it's very nature, spells an end to this equation of a particular piece of software for a particular application, which is the *identity* of propietary software, and it's raison d'etre.

In closing, while it appears that proprietary software firms have a rather bleak future, there appears little to worry about for those who do in-house software development, and even if FOSS is used more and more in in-house development the net result only partially contributes to FOSS because most of the resultant programming remains in-house and effectively proprietary. And this will likely not change unless the business model itself changes to something which is not based on data, and how that data is acquired, accessed and utilized.

Reply Parent Score: 3