Oh. My perferatory "hate to say it" was more along the lines of, hmm, I'd guess you could call it claims about the possible superiority of open source as a method of managing software projects. there are a lot of reasons why people advance free and open source software (FOSS). I'll just focus on a couple, people will have to refresh my memory as to other reasons that pertain to the *quality* of the software.
1. with enough eyeballs, all bugs are shallow
This view is no longer common, but was once the rallying cry, especially in the security community. Here, the idea is that software is so damn complex, based on so many layers of abstraction and interconnection between systems such that you can never dig your way down to a final arbiter of all that abstraction, that you will inevitably have bugs. It's just the nature of the beast.
Alas, so complex = not enough people could be hired to test the shit out of the software - never mind the scaling problem. With software, it's one thing to test on a separate testing server with, oh, if you're lucky 100 users. It's quite another thing to release it to "the cloud" where millions of people hammer it every day on a shit ton of operating systems with a crap load of different browsers with different security patches and then a whole slew of other software. Damn unstable environments, where no two uses brings the same code on the machines to the table to download and use yours.
the idea *was* that you could push out beta products and have users test it. Here, the focus is on *working* software, not developing in a corner somewhere until you have it fully featured and working perfectly. Release with all the problems and then users will test and report the bugs.
With enough eyeballs, all bugs are shallow.
didn't work out that way.... firstly, because people didn't want to mess with buggy software. secondly, because people just didn't want to be bothered to report the bugs. thirdly, because even if they might have been willing to do above, there is little by way of explaining this to ordinary users as to why they *should* help test beta &tc software.
2. another claim has to do with shedding the layers of useless management.
you don't need managers, product development, marketing, etc. because those people are ususally there to get people to do work they would otherwise rather not do (managers), plan out and design software full of features that users don't actually want (product management), plan and run projects to keep them on time and on budget (project managers), and explain to people why they need yet another useless upgrade when, in fact, they really don't (marketing).
under the layers of this sort of management, with drones working for soulless corporations, you can't get good software, by definition.
At 06:51 PM 6/25/2012, michael yates wrote:
>Sometimes you have to accommodate people in terms of the software you use.
>I edit book mansucripts as Editorial Director of Monthly Review Press.
>Word is the standard format so far for submitting proposals and then the
>first draft of the book. So if you are a microsoft hater and refuse to use
>word or a computer expert and prefer some other program, then what am I to
>do if your program and mine aren't compatible? If I use Track Changes in
>word to do my edits, and your program can't do this, then I have to waste
>such a shitload of time accommodating you when it should be you who
>accommodates me. I mean I really like wordperfect, but I had to learn to
>use word since most people do. We can't be purists all the time. I don't
>care to shop at Wal Mart, but I have been in places where I didn't have
>much choice. You need a pack of underwear, and you get them where you can.
>
>
>
>
>___________________________________
>http://mailman.lbo-talk.org/mailman/listinfo/lbo-talk
-- http://cleandraws.com Wear Clean Draws ('coz there's 5 million ways to kill a CEO)