I'm wondering about two things here.
First - in cases where proprietary software is used, its generally the case that the company has to develope its own middle-ware product for intergration into their systems (or buy a boxed one), either way, its part of the cost of the product. Is that being used in this calculation, or is that considered a seperate "product"? For an open source solution, the better option is probably just to modify for the source to integrate the product, which would be more cost for the product itself, but you miss the middle-ware costs.
Second - is this looking at only the "cost" or factoring in time as well? If you're having an integration problem (even if you are a big enough company to complain loud enough to get say, Microsoft, to implement a fix) you're certainly not going to get it over night, or next week. In smaller companies, this isn't an option at all. If you look at time waiting for Microsoft to probably never fix the problem, versus devloping and submitting an code change for an open source product, I'd bet the amount of time actually used is significantly less than the Microsoft turn-around. Sure its a higher development cost to the company, but it will save them many hours getting a fix to their workforce in a faster time frame.
Slashdot Mirror
I'm wondering about two things here. First - in cases where proprietary software is used, its generally the case that the company has to develope its own middle-ware product for intergration into their systems (or buy a boxed one), either way, its part of the cost of the product. Is that being used in this calculation, or is that considered a seperate "product"? For an open source solution, the better option is probably just to modify for the source to integrate the product, which would be more cost for the product itself, but you miss the middle-ware costs. Second - is this looking at only the "cost" or factoring in time as well? If you're having an integration problem (even if you are a big enough company to complain loud enough to get say, Microsoft, to implement a fix) you're certainly not going to get it over night, or next week. In smaller companies, this isn't an option at all. If you look at time waiting for Microsoft to probably never fix the problem, versus devloping and submitting an code change for an open source product, I'd bet the amount of time actually used is significantly less than the Microsoft turn-around. Sure its a higher development cost to the company, but it will save them many hours getting a fix to their workforce in a faster time frame.