On Sat, Sep 24, 2011 at 7:52 AM, Rahul Sundaram <[email protected]> wrote: > The goal of sticking to a single release is essentially because it is a known > quantity and that is very important when you have hundreds or thousands > of systems or even a few critical ones. Custom in house applications are > often written in enterprises and upgrading can break them in non obvious ways. > Unless the new release offers something that is compelling enough to think > about, mere inertia is enough to not change anything.
So far, enterprise was forced to upgrade their applications when they replace their aging hardware and the new hardware setup is not compatible with their old Apps. With Virtualization helping them, they can indefinitely postpone upgrades by maintaining the old setup even in new hardware. With this kind of inertia, even a 10 year support-cycle will not be enough for them. So why eliminate Ubuntu based on this criteria? On Sat, Sep 24, 2011 at 11:40 AM, Arun Khan <[email protected]> wrote: > On Sat, Sep 24, 2011 at 4:36 AM, Kumara Guru <[email protected]> wrote: >> If the enterprise wants a 10-year locked-down deployment even for an >> Open Web Apps stack, that makes me really sad. >> I was under the impression that only proprietary, commercial products >> lock down their customers and make upgrades a costly affair. > > As a general practice, upgrades (even .xy releases) need to be > validated - be it FOSS or proprietary. Validation costs money. I am not denying that. When designing apps around an open platform, I expect costs be less than intimidating to comfortably go for periodical planned upgrades. My only contention is that a *Web application stack* -- of all things -- should be on a shorter upgrade cycle. 10 years is a ridiculously long time by Internet standards. This is especially true for customer-facing applications like an eCommerce portal. Yes, it would be highly risky to make unstable upgrades to an eCommerce portal as any Failure due to it has direct financial consequences. However, if you shy away from even making significant upgrades in reasonable time periods, your App will be a dinosaur caught in a time warp. Until like 4 years ago, cloud strategy was largely unheard of, and many enterprise Web Apps had a not-so-great design for scalability with either a huge wastage of computing resources or an abysmal resource pooling. But now, significant cost savings can be achieved with a well designed cloud strategy (Disclaimer: I am not pretending to be a expert on cloud computing -- merely stating the obvious). If your eCommerce portal had no architectural redesign for the last 5 years and will continue to be so for the next 5 years, your are pretty much limiting the number of customers you can do business with. It would be unwise not to expect a better architecture to surface in much less than a decade from now and resisting the change at that point, will actually be a financial drain. Regards, Kumaraguru _______________________________________________ ILUGC Mailing List: http://www.ae.iitm.ac.in/mailman/listinfo/ilugc
