| PLEX86 | ||
winscape 2328snip winscape 2329 Yes. Especially applications. App people have absolutely no idea about how to solve a computing problem for the general timesharing public user. Thus, a fix to underlying software that works for... Not at that company, no. Correct. So, you're telling me that experience developing (or maintaining) application-level software gives essentially zero insight into the "work flows that have to be done in order to ship one product." That seems strange to me. I agree that writing applications software doesn't necessarily teach a person anything about writing compilers or operating systems or developing hardware. Maybe I don't understand what you mean by "work flows". I would have said this had something to do with the interaction among the sales-marketing people who (in my experience) have significant input into specifying what gets built, the technical people who do the building, the people who test what gets built, the customer base, and so forth. I would also have said that while this interaction wouldn't be exactly the same in a large organization writing o-s-level software and in a small organization writing application software, it also wouldn't be totally different -- in particular, in both organizations, there would be tradeoffs between what the technical people might want (enough time to do a really good job, including time to do a total rewrite if that was indicated) and what the marketing people might want (earliest possible ship date). Not exactly, but something along those lines. This was a company that made mainframes that were intended to run IBM mainframe operating systems. Getting IBM's code to work on the company's hardware required applying what I'll call "local modifications" to IBM's code. Customers would buy our hardware and IBM's software, and we would ship them "bug fix" tapes with our local mods. At that time IBM was sending out upgrade bug-fix tapes about once a month; the group I worked with was supposed to apply these upgrade tapes to our baseline system, reinstall any local modifications that were wiped out by the upgrades, and test the resulting system. Most of our testing was on a virtual machine (under VM-370), but sometimes we tested on real hardware. I've forgotten how we packaged the results for customers, but the primary goal of our group was to figure out what had to be done in order for them (customers) to install IBM's upgrades without breaking the local mods they had previously received from us.
I think even someone suffering from PCitis might object to having to do a reinstall after every 100 hours of use. That's, what, every 2.5 weeks given 40 hours a week of use? However, yeah, an operating system that needs to be reinstalled for no reason other than accumulating bitrot .... -- B. L. Mbuttingill ObDisclaimer: I don't speak for my employers; they return the favor.
|
||||
Alt Folklore Computers from Newsgroups The #1 Usenet Provider on the Internet
|
||||