It also makes me wonder whether this is a 'weeding' course where having too many passing grades would cause the school to have too many sophmores going forward. If they've intentionally over-admitted freshmen in order to filter out the best ones for the sophmore class, having a group that significantly increases the number of passing grades in one of the key killer classes might be a serious administrative problem.
Given that a curve could still fail students with numerically high average grades, folks tend to get far more bent out of shape by scaling a %85 to a D- than by scaling a %32 to a B+. A broad based and effective facebook study group could result in far too many numerically passing grades and need to be suppressed...
Microsoft may not get it right, but they certainly don't want to spend lots of money supporting broken products. As best I understand open source business plans, the most successful sorts of open source products (at least in terms of supporting jobs and revenue) would involve very capable products that are very hard to use properly. Keeps the consulting revenue up and keeps existing customers paying folks to support and customize them. Not an unreasonable strategy...pretty much the ultimate extension of how HP makes money on ink-jet printers...sell the main product cheap (in this case free) and make sure that the customer needs to keep coming back to you for supplies (or in this case support and customization). With the right mix of products this can be made to work.
Sounds to me as if the way to make valuable open source products is to create a product that is very difficult to setup (thus producing much consulting/support revenue) but very powerful once you've got it going. If you can find something that requires extensive customization you're probably on the right track. Easy to configure products that are readily usable by everyone don't contribute usefully to the open source community's economic well-being (as they'll just be used by non programmers and won't fund any developers by way of support contracts)...Interesting... This suggests that the income proposition for opern source products is almost backwards from that for closed source. A closed source commercial company wants to provide the product pretty much ready to go and doesn't want to provide extensive after sales support. An open source company wants to release products that require extensive support as paid contract work (and this sort of product enriches the entire open source community...at least as long as the end result of the customization is quite valuable)...
Slashdot Mirror
It also makes me wonder whether this is a 'weeding' course where having too many passing grades would cause the school to have too many sophmores going forward. If they've intentionally over-admitted freshmen in order to filter out the best ones for the sophmore class, having a group that significantly increases the number of passing grades in one of the key killer classes might be a serious administrative problem. Given that a curve could still fail students with numerically high average grades, folks tend to get far more bent out of shape by scaling a %85 to a D- than by scaling a %32 to a B+. A broad based and effective facebook study group could result in far too many numerically passing grades and need to be suppressed...
Microsoft may not get it right, but they certainly don't want to spend lots of money supporting broken products. As best I understand open source business plans, the most successful sorts of open source products (at least in terms of supporting jobs and revenue) would involve very capable products that are very hard to use properly. Keeps the consulting revenue up and keeps existing customers paying folks to support and customize them. Not an unreasonable strategy...pretty much the ultimate extension of how HP makes money on ink-jet printers...sell the main product cheap (in this case free) and make sure that the customer needs to keep coming back to you for supplies (or in this case support and customization). With the right mix of products this can be made to work.
Sounds to me as if the way to make valuable open source products is to create a product that is very difficult to setup (thus producing much consulting/support revenue) but very powerful once you've got it going. If you can find something that requires extensive customization you're probably on the right track. Easy to configure products that are readily usable by everyone don't contribute usefully to the open source community's economic well-being (as they'll just be used by non programmers and won't fund any developers by way of support contracts)...Interesting... This suggests that the income proposition for opern source products is almost backwards from that for closed source. A closed source commercial company wants to provide the product pretty much ready to go and doesn't want to provide extensive after sales support. An open source company wants to release products that require extensive support as paid contract work (and this sort of product enriches the entire open source community...at least as long as the end result of the customization is quite valuable)...