* Serguei Chabanov; <chabanovs@runbox.com> on 01 Nov, 2002 wrote:
On Friday 01 November 2002 01:19, gilson redrick wrote: Now there will be even more distributions from the same company. And how many more linux distributions there are since 7.3? All born from the same dissatisfaction with the existing ones.
My argument is this: we are actually witnessing linux distributions succumbing to the bazaar model, not the other way around.
Any comments?
From the business perspective this is a core issue, I'd rather have my employee to deliver what they are suppose to in a timely manner with
Well although I was trying to avoid this thread as it is going towards a general advocacy discussion, I could not resist to jump in. The problems I see relates to the whole Linux community at large. 1) The majority wants to have a stable working environment, either be it the operating system lying under or the applications that are running. professional quality. Their problems related to the Operating system and the applications they use can not be an excuse. If they are going to spend the time ,which I pay for, in fixing problems related to the use of their choice of operating system or application, for me that is a valid reason to ask the person to leave the job. It may sound harsh but this is reality my time my money. If that person wants to play with a different distro, operating system, application fine he can do so in his leisure time. As I see it this the area where companies have problem. Linux in general is thought of as free beer which makes it a perfect alternative for replacing operating systems and applications in an effort to reduce costs in a continously shrinking economy. On the other hand this approach brings the disadvantages along the unstable working environment due to the "release early, release often" approach either the software is missing critical features or is in an unstable state. "Release early, release often" is a good way to enable the users to have the possibility of catching the missing features in a very short term. I see this approach as a (alpha/beta) era which can not be adapted to the business environment. Not to mention that these releases are tested by anyone who wants to try it so the quality of beta testers is questionable. I would expect a beta tester to provide a clear picture enabling the developer to reproduce the same error. Well this is not the case in the majority of events. Just have search to the achieves of the suse-linux-e mailing list you will see plenty of examples. "XYZ does not work help me" will be very common, which does not help at all. Since the developer does not have enough information he is unable to provide a working solution if it is a real bug. Consequently the quality of software is low. At this very point the distributor, can not be held responsible as he has no control over the quality of software development. That is when support agreements come into effect if you as the business are willing to pay for the expenses of making the software work then things may improve. Suddenly people start shouting "hey Linux was free." So we go back to the very first step. 2) The SuSE business packages are an answer to that in my point of view. If you want as the business to have the job done while trying to cut costs then you will reach to concensus. If I have invested in converting my Exchange Server to Open Exchange I do expect it to work as advertised and that is the reason I am paying the bill. If it would be like the general distribution than I would start arguing. So one way to enter into a new market is to understand your weaknesses and understand the needs and desires of your customer ( and potential customers) and adjust your products accordingly. You need to differentiate the customer and the product in order to increase your market share (and actually wallet share). Having the biggest market share means nothing to me if that share has not generated profit. I would rather have a small market share but a bigger wallet share. So as a Linux oriented company you need to make a portfolio analysis to see "life cycle stage" "market attractiveness" "relative market share" and decide what to do with your "Cash cows" "stars" "loosers" and "question marks" [1] What is a "question mark" today could be your "star" I would assume SuSE has already done this analysis and moving accordingly to the plan. Yet it does take time to move a "question mark" product to a star in a rapidly growing market/environment. So overall I would say SuSE is moving towards the Cathedral but not as fast as I would like to see it. Sorry for a long one but this is how see things. [1] Strategic Market Planning Robert. J Hamper L Sue Baugh -- Togan Muftuoglu Unofficial SuSE FAQ Maintainer http://dinamizm.ath.cx