Xref: utzoo comp.org.eff.talk:2509 alt.privacy:741 alt.censorship:2435 Path: utzoo!utgpu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!think.com!spool.mu.edu!agate!violet.berkeley.edu!turrell From: turrell@violet.berkeley.edu (David Turrell;;;;GQ79) Newsgroups: comp.org.eff.talk,alt.privacy,alt.censorship Subject: Re: Prodigy charged with invading users' privacy Message-ID: <1991Jun3.055353.24642@agate.berkeley.edu> Date: 3 Jun 91 05:53:53 GMT References: <1991May24.225008.6510@unlinfo.unl.edu> Sender: root@agate.berkeley.edu (Charlie Root) Organization: University of California, Berkeley Lines: 57 >From: dogear!kharma!dave@isc-br!tau-ceti >Date: 27 May 91 08:30:21 GMT >Message-ID: > >I think I have a handle on what has happened. Prodigy blew it, not from a >point of spying on people, but on stifling an honest, open response to the >public's concern. In other words, the worst thing they did wrong was a PR >gaffe of monstrous proportions. Their PR *used* to be good. I know I was intrigued by their ads and what else I heard about them before the stuff hit the you-know-what. I couldn't ever believe they were creatively spying on their customers, perhaps because I've worked for big brotherish companies and know how much a waste of everyone's time that would be. In the first place, it's not necessary. A computer service has a pretty good idea of where you fit on the scale of fundamentalist to far-out by the services you use. All my service, Compuserve, has to do for market reasearch is find out which forums I have joined, which databases I query and which CB channel I hang-out on. Never mind that they can read my e-mail if they have a mind to. Secondly, the demands on Prodigy's systems would be enormous, if they decided they were going to start uploading a large amount of data from their subscribers' machines. Care to calculate the added amount of processor time, number of disk seeks, etc. needed to upload and stash on Prodigy's system all the secrets that its subscribers have got stashed on their systems? So that does leave only the fact that people were willing to believe the worst about Prodigy because Prodigy had a major talent for p*ss**g people off. >[...] > >That isn't to say I don't think that Prodigy *COULD* be doing something funky >in the past. There are a *lot* of unanswered questions left over from when the >fiasco first started. I will always wonder what they really had in mind with >STAGE.DAT, probably never will really trust them, and do not recommend that >anyone with sensitive data on their hard disk call them without taking >precautions. Gee, does this make me sound paranoic? And you were sounding so sane up until now. I haven't heard much about what STAGE.DAT does, other than wind up with a lot of data no one wants it to have. Given Prodigy's use of graphics, it must contain, among other things, video display refresh buffers. The name of the file evokes a theatrical stage with its scenery flats that are lowered and raised as needed. STAGE.DAT probably has several refresh buffers whose starting offsets are loaded into the display device driver whenever Prodigy undergoes a change of "scenery". This might also account for some of its sloppy behavior. Graphics programmers are used to being under pressure to do a buffer update between scans and they often regard niceties, such as zeroing out allocated RAM prior to use, or using sparse matrix techniques to write to a file only the RAM they use, as sacrilegious. I hope Prodigy will give a better explanation sometime. -David Turrell