[ILUG] Wow! What a response to an almost throwaway comment (was: Naive ... (long))

[Brian Foster]

 I'm impressed:  An avalanche of c.80 messages with two
 thread forks and c.½MiB, all in c.48h, to what was
 essentially an aside, albeit a LOUD one.  And no
 flaming (or at least none which was obvious (to me)).

 Below are my comments and clarifications on that one
 issue; especially on what was said that is directly
 relevant to my off-hand comment (as of c.noon CEST,
 23rd June).  I've tried to present the context and quotes
 fairly, albeit not necessarily in chronological order (and
 hence I've also ignored the threading).  Please accept
 my apologies if I've misrepresented anyone.  And please
 note I'm only replying to points of my own selection,
 so there is quite a bit I'm ignoring (generally, I've
 nothing to say/add in those (ignored) cases).

 First, however, I'd like to thank everyone for the
 answers and suggestions to the questions I asked.
 I'm still crunching and researching what was said.
 I presume I'll have some follow-ups later on?

 Back to the avalanche ....   What I commented was:

   “(I note that, according to the (downloaded) contract,
    Linux is not (listed as) supported.  That does not
    worry me (should it?), except if they insist Windross
    must be used for diagnostics / upgrade / installation
    or whatever, which is a non-starter: I *** REFUSE ***
    to run M$ shiteware.  100% ABSOLUTE REFUSAL!)”

 The concensus seems to be the Windross requirement is,
 technically, not relevant.  I myself suspect it is just
 a weasel clause, and might imply the support is of the
 mickey mouse variety.  (That is, follow a script, and
 give up as soon as the script isn't followed.)

 No one, however, seemed to react to it being a contractual
 requirement (albeit Josh Glover mentioned a related
 issue).  With the usual caveats (IANAL &tc), that may
 be an issue in some improbable future (e.g., a dispute).
 I didn't say so, but the contact could be rather sloppily
 worded:  E.g., it lists specific Windross versions (but
 not, interestingly, Vista), but fails to mention anything
 about the installation being any of legal (Josh's point),
 maintained, updated/patched, &tc, albeit it apparently
 must be used (and hence, presumably, functional).  OTOH,
 it is my responsibly to ensure the provided link is not
 used for spamming, &tc, implying I must make a reasonable
 attempt to keep things secure.  Since I would do so
 anyway, trying to keep things secure is certainly not
 an issue!  (The Mac is an alternative, but is even less
 precisely defined; a Mac running Linux _might_ actually
 meet the terms of the contract!?)  I think these O/S
 clauses in the contract are more amusing than important.

 And full credit where credit is due:  David Golden pretty
 much hit the nail square on the head as to one of the
 reasons for my refusal:

   “[... M]aybe, not a windows user?

   “Windows has always been a clunky-looking and
    unfamiliar OS for me -  I basically switched to
    Linux-on-PC from Linux-on-Amiga from AmigaOS-on-Amiga,
    Windows was a sideshow that appeared installed on the
    first PC I bought, though not later ones.

   “So I haven't had a windows partition on a computer
    I own since about 2000, and most of the rest of my
    family now use Macs.”

 Paul Jakma also alluded to essentially this same point,
 multiple times, such as:

   “There are at least a few list members who have
    no use for windows and use some flavour of Unix.
    Not out of zealotry, but because of preference, habit,
    employment and other such sensible reasons.”

 and:

   “If one has no use for Windows, what's the point in
    keeping it??”

 In my case, after using whatever systems I could obtain
 time on (OS360 with punch cards (obsolete even then),
 RSTS-11, TOPS-10, and others), I finally got access to
 Unix (32V) on a VAX/780; that would have been c.1980.
 I've been using and programming *ix systems ever since
 (and various embedded microkernels, all of which have had
 a very strong *ix flavour (which, as Paul observed, is
 often the case for embedded routers)).  For all intents
 and purposes, that start with *ix was before the PC
 and M$-DoS, much less Windross; so like David said,
 “Windows was a sideshow ...”.  Or as Paul said,
 “... because of preference, habit, employment ...”.

 And hence full marks to Paul, who also correctly deduced:

   “Given Brian stated he refused to run MS software,
    one can presume he would delete any installation of
    Windows and re-use the disk space for more productive
    purposes..”

 For years the only thing I ever did with DoS and
 Windross was overwrite it with various flavours of *ix.
 That record broke at my last job, where Windross was on
 every desktop, albeit for the most part, I used it only
 as an X-terminak to the (Linux) servers.  (The company
 procedures required some use of M$-Worthless, which only
 served to inhibit me from writing papers, since I found
 it difficult and clunky and awkward to use.  I wasn't
 the only one annoyed at the M$-Worthless policy, but it
 wasn't until 2.1 that OO.org could be even be imagined
 as perhaps an allowed alternative/replacement sometime
 in the far future.)

 Timothy Murphy started the avalanche sliding, suggesting
 (at different times) that perhaps there is some degree
 of both fanaticism and masochism in my refusal, and
 also offered a serious of stories where he allegedly
 needed to use either DoS or Windross.  Several people
 objected, both to the fanaticism and masochism (dubbed
 “zealotry”), and to the alleged need in the stories
 to have Windross installed.  I've only one point to add:
 An assumption that at some time I needed to use DoS or
 Windross and made life hard for myself by not doing so.

 That assumption is false.  I can only recall two incidents
 over the last >25 years where I know, or suspect, I
 _needed_ either.  In one case, to update the FLASH-BIOS
 from a prototype to the production version (which was
 done by booting a floppy, so it didn't matter the laptop
 in question was all-Unix (and so I'm only guessing the
 floppy used DoS)).  And the other case was to run a
 microcode design/verification tool; this was done by
 borrowing time on another lab's Windross machine (with
 full permission and some assistance).  I would have had
 to do that anyways, since I wasn't using PCs or Linux
 in that job (and hence have no idea if Wine would have
 worked or not).  And as it so happened, whilst the
 microcode did contain a bug, it wasn't the problem:
 A unconnected pin was the mistake.  So I didn't need to
 use Windross at all, but I didn't know that at the time.

 I did use Wine, and also SSH'ed into Windross, in
 my last job, but again it's not clear I _needed_ to.
 In this case, doing so was simply the fastest way to
 again do some verification (albeit this time of C and
 C++ code, not microcode), and had the questionable
 advantage the verification was done with the same
 not-always-reproducible installation(s) used for
 production &tc.  (It turned out at least one of
 those installations was also not properly backed-up!)
 And anyways, when SSH'ing to Windross, I used Cygwin
 (the company's policy), or MingW (my own preference).

 Except, maybe, for the FLASH-BIOS updating, none of these
 cases are those Tim postulated, which is fair enough,
 as he was only speaking of his experiences.

 On the FLASH-BIOS updating, an assumption seems to have
 been made which isn't (in my experience) true:  That you
 need to update.  The very useful “If it ain't broke
 don't fix it” applies, especially since, on *ix systems,
 the BIOS is mostly not relevant after early-stages boot.
 That is, ignoring any non-technical issues (e.g., support
 or company policy), if you can boot, the BIOS is Ok.
 (APM and maybe other features mean this isn't as true
 as it used to be, which is very unfortunate.  Note said
 features can be disabled.)

 Paul also pointed out, for the FLASH-BIOS story,
 one option of many is to “Write your own firmware
 update[r]”.  Whilst that is presumably beyond the
 ability, interest, or commitments of many people, it
 is, in my own case, plausible, since I _have_ written
 firmware updaters (plural), albeit not for PCs plus some
 other caveats (e.g., in a well-equipped lab).  However,
 without stuff such as a lab, I'd be extremely unlikely
 to go that route if I had the postulated problem.

 Rory Byrne made several good points (and incorporated a
 valuable update from David), including:

   “Some people would rather make an up-front investment
    of time, money, or whatever to change their situation
    rather than just deal with it.  Buy Linux [friendly]
    hardware, set up SIP or IAX rather than use Skype,
    etc.  I'm not advocating one path over the other -
    just saying that there are two paths and there are
    travellers going down both of them.”

 A minor quibble, I'd actually say “... there are
 multiple paths ...” (not just two).  John Madden put
 it like this:

   “... if they absolutely require Windows, then politely
    explain that you'll be going to a competitor because
    of this.”

 and:

   “It's "the way it is" because people don't complain
    about it.  If people complained and / or went to a
    competitor, things would change. Look at Dell - it's
    taken years, but they're finally offering Linux as a
    pre-installed option, because customers want it.”

 I do complain (and also compliment).  I did once have
 a problem with phone support, but since the machine was
 completely dead (not even the LEDs were lit), it was
 self-evident the software didn't matter.  Credit to the
 person on the other end, who got the point and authorised
 the repair under warranty.  So full marks to both Rory
 and John (and, again, to David, who also made the point).

 Jimmy Tang asked:

   “so the question remains, how many linux users
    here who claim that opensource is great and strongly
    disagree with using windows because its a tax that
    manufacturers impose on users also use things like
    binary blobs (un)knownlingly in their kernels to get
    hardware that the linux user has bought[?]  a good
    example of this is nvidia ...”

 There are too many caveats (riders) attached to the
 question for any such self-selecting survey to be
 meaningful.  It's not too clear what the best set of
 questions might be, but a start could be “How many
 Linux users strongly avoid using Windows?” and “How
 many Linux users use binary blobs?”.  To which _I_
 answer “yes” and “no” (despite having an Nvidia card),
 or more accurately, “yes” and “not any more” (my dead
 Epson printer's Linux host software used a binary blob;
 my new HP printer's Linux host software is from HP and
 is all on SourceForge).

 The HP printer is currently dead, and being repaired
 under warranty.  Linux wasn't an issue, partly because
 it was another total-death case, but mostly because HP
 does provide software for Linux.  However, the (e-mail)
 support isn't geared up to help with Linux (and admits
 this), and so whilst (in my current on-going experience)
 they aren't hostile and don't give you the runaround,
 they also aren't much help.

 To summarise:  Prior to getting my machine many years
 ago, I had no reason to believe I'd either want or need
 Windross; and anyways I can't use it effectively; my
 guess was correct; and combined with other reasons (such
 as, but not limited to, concerns about M$'s policies
 and ethics), I continue to fail to see any reason to
 have or run Windross (excepting, e.g., company policy,
 a situation I hope to never encounter again!), despite
 the odd support hassle (which is a cause for complaint).

cheers!
	-blf-
-- 
▶ ▶  I AM CURRENTLY LOOKING FOR A JOB!  ◀ ◀ | Brian Foster
Experienced (>25 yrs) software engineer:    |        Montpellier, FRANCE
 • Unix, Linux, embedded, design-for-test;  | Stop E$$o (ExxonMobile)!
 • Software/hardware co-design, debugging;  |     http:/www.stopesso.com
 • Kernels, drivers, filesystems, &tc;    Résumé (CV) & contact details:
 • IDL, automated testing, process, &tc.   http://www.blf.utvinternet.ie