Archived from groups: alt.comp.periphs.scanner,comp.periphs.scanners (
More info?)
On 8 Sep 2005 19:44:24 -0700, "Noons" <wizofoz2k@yahoo.com.au> wrote:
>> >But to write something
>> >as complex as Vuescan and make it look the same and operate the same
>> >in all three environements is a major achievement.
>>
>> Please forgive me, but that's just nonsense.
>
>There is nothing to forgive, I do respect anyone's opinions
>no matter how different they are from mine.
Yes, but there are much more civilized ways of expressing that. Even
though it's not an excuse I was just pressed for time but still felt
uncomfortable with the term. So, thank you for understanding!!
>> Vuescan is not complex. It's mind-numbingly simple, assuming a
>> competent software engineer. Cross-platform development has been going
>> on for decades and has absolutely nothing to do with Vuescan problems.
>
>Beg to disagree. If handling scanners in multiple platforms
>was simple, there would be a LOT more programs like Vuescan.
>There is only one so far that covers that many scanners in
>that many platforms. There must be a reason for that, it's not
>like there isn't a demand for such programs.
I think the reasons for that are much more mundane. Specifically, two
key ones: market share and native software.
Scanner programs are still a niche market, relatively speaking,
meaning there's no sufficient volume which would make it attractive
for the "big boys" to enter.
But, perhaps more importantly, scanners already come with native
software which is more than adequate for vast majority of users. There
is usually a quick-and-dirty "auto" mode for "civilians" and after
turning all the settings off the "pros" can still get their raw scans.
Indeed, the Vuescan author himself once wrote that he saw these native
programs which came with the scanner as his competition, not
SilverFast.
A scanner program at its most basic is simply data acquisition. And
that's very straightforward. Especially, now that the hardware
interface has been standardized (i.e. USB with even FireWire on the
decline) instead of a myriad of custom "solutions" like those parallel
port scanners of a few years ago.
What's left is image editing (which in my opinion doesn't even belong
in a scanner program but the "civilians" need it). And those image
editing routines are well known and readily available.
So given all that, making a single program access different scanner is
really very elementary.
>> To combat this some methodologies actually require programmers (after
>> an initial design phase) to literally throw away *all* notes and start
>> from scratch. The idea is you can start fresh with the new knowledge
>> but without the burden of an existing design.
>
>Sure. But this is not a software house with standards,
>methodologies, etcetc. This is a single man band, doing more
>than a lot of companies with all those in place.
Yes it's a one man show but that's not the issue.
By the same token, a user could say: I'm not a big company but just a
little guy here scanning my family photos, can I get your software for
free?
Not much sympathy for that now, is there?
Conversely, whether the software is made by one man or a
multinational, in the context of a marketplace it lives or dies based
on what it does or doesn't do.
And there Vuescan fails miserably even if we do cut it some slack.
Even a one man band can check if the software actually scans before
unleashing it on the unsuspecting public!
And when that public complains, instead of fixing it he sends emails
to his users telling them "if you don't like it, delete it" or tells
them they've been "blacklisted" because they *justifiably* complained!
>point: Silverfast, from a much bigger stable. Yet feature-wise,
>Vuescan leaves it for dead. Sure, Silverfast has a much more
>polished interface for first time users.
And that's the key! SiverFast's target audience are "civilians" who
only want a big, single "auto everything" button.
Vuescan, on the other, hand is "scattered" (a reflection of a badly
thrown together "user interface") oscillating between "auto" and
"technical".
What is the point of simply listing many "features" if they simply
don't work, or worse cause damage?
>I much prefer Vuescan's option: listbox, pick between "none",
>"light", "medium" or "heavy". Call me crazy, but it is clear
>and concise...
That's personal preference. By the same token, others may find such
description simplistic and would want a numerical display, or
whatever. But that's not the point...
What I'm talking about is something completely different. Take the
"display refresh" or whatever it's called in Vuescan. I'm referring to
entering a value in the setting and when after each keystroke the
whole screen redraws. Typing too fast causes keystrokes to be
"swallowed" without *any* feedback! So, instead of exposure "50", one
would end up with "5".
Yes, there is a "secret" option to turn the "screen update" off. But
that's a cure worse than the disease!! Now the display has nothing to
do with actual settings currently active. That's positively crazy! And
certainly brakes every UI guideline regarding ergonomics.
>Note that I am not denying there are problems with the UI.
>One that you mentioned before and I particularly dislike
>is the quaint way it changes the appearance of other tabs
>when I change an option in the current tab. If I go and check
>why, it makes sense it did so. But it is disconcerting when
>one starts using it and it does that!
>
>Still, I can't afford the time to think about it and contact the
>author with alternative ways of handling those issues.
>Maybe you could do that? Or have done so and got nowhere?
I have no interest in Vuescan, whatsoever.
The author is notorious for his short temper and abusive outbursts. In
the three years I've been here he's still the only one to have
actually screamed obscenities...
I gave Vuescan early on because NikonScan did not have Kodachrome
option for my scanner at the time (LS-30). Now it's an LS-50.
Back then it struck me that Vuescan did not have individual Analog
Gain setting. This caused a saga you can read in the archives:
The author's (arrogant) response was "You don't need that!"
(Since then I learned that's his favorite response. Apparently,
initially, Vuescan didn't even have a preview window, and when people
asked for it, that's the response they got... :-/)
Anyway, I did "need that" and the author went through several
contradictory statements starting with "I can't implement individual
AG because it would upset my color balance" and ending with
"implementing individual AG is elementary". Hmmm...!?
That didn't sit well with actual users and they sheepishly started
saying they would like individual AG too, so the author "challenged"
me to come up with a slide to "get to the bottom of this".
I promptly posted one *and* also included a scan where Vuescan
performed well to show both my objectiveness and to prove there was
nothing wrong with the scanner or the installation. (BTW, if you check
the archives the images are still available!)
To make a long story short (too late! ;o)) the author said fixing this
image would be easy. I asked him to provide specific settings (so that
I can't be accused of sabotage and give him a chance to prove me
wrong).
Unable to provide them, he exploded with a tantrum and abusive
language.
When he calmed down he implemented individual AG which Vuescan users
begged him to do for months. The thanks I got from those very users is
that they too started hurling abuse... Go figure...
Anyway, don't take my word for it. It's all in the archives!
>> author just got into ever more trouble. A classic symptom of this is
>> when bugs proliferate and never go away and perennial unreliability.
>> Sounds familiar?
>
>Does it ever! Having been on the software business for many years,
>I know exactly what you talking about.
>But I disagree that there is perennial unreliability there.
>It has crashed my system the grand total of twice (and locked up once)
>since I started using it. Cripes, Firefox has done so more times
>than I care to mention and I keep using it! See what I mean?
My Firefox hasn't done that... yet... ;o) but I'm afraid Vuescan *is*
unreliable when I read all the complaints after each new release.
Now, being in the business, I suspect you're using the software
"defensively" i.e. not upgrading just because there's a new version,
as well as keeping a "known good version" backed up "just in case"
whenever you do decide to upgrade - after others have beta-tested it
for you... ;o)
Also, it may very well be that employing such a strategy you have
found a version which serves your requirements, but that's only a one
person's experience.
If we examine Vuescan in its entirety and objectively then the only
conclusion is that it *is* perennially buggy, although you or some
others may have found a combination which satisfies your/theirs
specific requirements.
There's no contradiction between that subjective, narrow definition,
and a generic, objective definition of Vuescan as perennially buggy.
>> Not at the same rate and persistence as Vuescan's!
>
>Can't comment there. I'm on my first download of
>the product, so I don't have any experience on
>bug persistence.
Oh, boy! You're in for a fun time!!! ;o)
Anyway, that explains it! Do check the archives for a litany of
reappearing Vuescan bugs. If you plan to use it, it's certainly in
your interest to do that!
>> How do you explain the most pedestrian bugs like "broken cropping" or
>> "0-byte file scan", etc appearing in major version *8*? That's
>> inexcusable! It indicates massive incompetence and total absence of
>> any semblance of quality control.
>
>I'm at a loss here. "broken cropping"? It seems to work fine for me.
Yes, there were several threads, over the last 3-4 months where
cropping did not work. Did... Did not... Did... Did not... Etc...
>"0-byte file scan" also leaves me perplexed. Care to explain
>a bit more about this one: haven't hit it.
Yes, there was one version where the Vuescan would start a scan, and
then huff and puff, only to produce a 0-byte file. Many angry users...
When challenged recently I posted a partial list of various Vuescan
bugs. Considering you just downloaded it this should be "required
reading". Check the archives for the following message:
On Fri, 29 Jul 2005 20:51:05 +0200, Don <phoney.email@yahoo.com>
wrote:
Subject:
Re: Vuescan 8.2.24, what's new: "Significantly improved infrared
cleaning "?
>> I know some Vuescan fans will emotionally overreact but it's a simple
>> statement of fact when I say that I have never seen a program so buggy
>> and unreliable as Vuescan in my 25+ years in the business.
>
>Obviously you have not been involved with databases!
>(DAMHIKT...)
Ha! ;o) And double Ha! ;o)
From IBM mainframe's *logical* IMS (for those in the know the
"logical" bit is important!) to Microsoft's unilateral re-definition
of NULL! And just for the heck of it, another "Ha!" ;o)
Don.
Facebook
Twitter
Instagram