Why wouldn't it be appropriate? That approach is fairly common - Apple does it, Microsoft does it, Cygwin Unix does it, probably others do it as well.

They would presumably be looking only at their own software, and they could leave the old manual method available for those who choose it.

-- Jack Hamilton jfh@alumni.stanford.org Caelum non animum mutant qui trans mare currunt.

On Sep 1, 2010, at 12:57 , Winston Groenewald wrote:

> Perhaps the bigger issue is whether SAS Institute would (or should) consider > it appropriate to analyse your disk to determine what to update. > On 31 August 2010 21:27, Jack Hamilton <jfh@stanfordalumni.org> wrote: > >> Exactly. The update program should analyze what you have and present >> you with a list of what you can upgrade. Then it downloads only what it >> needs to. This nonsense of going through multiple pages of hotfixes to >> see which ones you need is just ridiculous. >> >> It could easily give you the option of upgrading only what you have >> licensed, or everything you have installed. It could create a mini >> software depot letting other users reuse what you've already downloaded. >> I don't see a need to prompt for site number or setinit - if you have a >> working version of SAS, the software can figure it out. >> >> The installation process is easier than it used to be, but it's still >> bad. I don't see that their new hotfix system has helped at all. >> >> >> >> >> >> On Tue, 31 Aug 2010 10:47 -0500, "Matthew Pettis" >> <matt.pettis@THOMSONREUTERS.COM> wrote: >>> I compare it against Linux-like package installs, with >>> apt/aptitude/rpm/yum etc., or PPM for perl... It just works. It knows >>> your versions, your dependencies, and it just does the right thing >>> quickly and efficiently. And those things are free! And yeah, it would >>> have to manage order numbers and whatever their term is for our unique >>> passwords, but that shouldn't be a big deal. >>> >>> I guess there are at least two ways to compare it -- to history, where >>> it is better than it was in say the 70s, 80s, or 90s, or to the current >>> state of other companies managing their software, in which it compares >>> miserably. Not having a history with SAS that far back, I don't have >>> the perspective of the former, only of the latter, and that is why I'm >>> frustrated. >>> >>> I expect better from a company whose product for which people pay (and >>> is not inexpensive, to boot). >>> >>> Matt >>> >>> -----Original Message----- >>> From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of >>> Peter Flom >>> Sent: Tuesday, August 31, 2010 10:35 AM >>> To: SAS-L@LISTSERV.UGA.EDU >>> Subject: Re: SAS and its horrible no-good very bad software management >>> system >>> >>> Ron Fehd wrote >>> >>> <<< >>> Oh, consider the alternative: >>> Individual hotfixes, >>> yeah, I know that pair is redundant. >>> >>> From where I sit >>> The current system is Sliced Bread >>> You want the Latest&Greatest? >>> -- like EG v4.3 -- >>> Get back to me tomorrow and I'll have the depot >>> And all the HotFixes issued up to end of last week. >>>>>> >>> >>> Well, consider how R gets updated: Twice a year, there is a new version. >>> You go to the website, click a button (one for Windows, one for Mac, >>> there's >>> other options too), agree to a couple options and bingo! In 10 minutes >>> or >>> so, you are done. >>> >>> SAS would, of course, need to add a spot to enter your authorization >>> code or >>> whatever, and it might take a little longer than R, since SAS is big, >>> but >>> why is it so hard? Why can't the authorization code automatically >>> determine >>> which parts of SAS you have licensed? >>> >>> I wonder how much Tech Support time is devoted to installation issues. >>> I >>> know that at one point (this was probably 2008) it took me an hour with >>> Tech >>> Support just to update. I doubt I am alone. >>> >>> Peter >>> >> >> >> -- >> Brevis esse laboro, obscurus fio. >> >> Jack Hamilton >> Sacramento, California >> jfh@alumni.stanford.org >> > > > > -- > Winston Groenewald > > --- An adult dove crashed full-speed into our glass patio door; death was > instant. Curious, I measured the wing area and weighed the bird while still > warm, gathering data for wing-loading. For this dove, it turned out to be > .87 square feet per pound. I imagine, however, the body and tail also > contribute a bit to lift. Applied to an average 160-pound human, the wings > would have to be about 38 feet in span, disregarding weight of the wing > itself. I haven't seen any angels in statues or pictures with near this size > wing, but Divine Assistance is probably another factor. ( > http://www.salon.com/tech/col/smith/2007/10/26/askthepilot251/index1.html) > ---